alsa driver for digigram lx6464es
hi all,
i have implemented the prototype of an alsa driver for the digigram lx6464es audio interface. audio i/o is basically working, although not very well tested, yet. the driver is available from my git repository [1].
features:
- i/o support for 64 channels via ethersound [2]
missing features:
- handling of xrun notification
- configuration of the ethersound network needs to be done externally
takashi, i would like to submit the patch to alsa soon ... how should i proceed? can you do a review of the code?
thanks, tim
[1]
http://tim.klingt.org/git?p=digigram.git;a=shortlog;h=refs/heads/lx6464es
[2] http://www.ethersound.com/
Signature · 5 lines
[email protected] http://tim.klingt.org I don't write music for sissy ears. Charles Ives
_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
At Sun, 22 Mar 2009 20:34:08 +0100,
Quoted 5 lines · Tim Blechmann
hi all,
i have implemented the prototype of an alsa driver for the digigram lx6464es audio interface.
Great.
Quoted 6 lines
audio i/o is basically working, although not
very well tested, yet. the driver is available from my git repository [1].features:
- i/o support for 64 channels via ethersound [2]
Quoted 6 lines
missing features:
- handling of xrun notification
- configuration of the ethersound network needs to be done externallytakashi, i would like to submit the patch to alsa soon ... how should i proceed? can you do a review of the code?
Simply post patches to alsa-devel ML and Cc to me (or vice versa). But I'll be on vacation from this week (Wednesday) until the Easter holidays. So, sooner better ;)
Thanks,
Takashi
PATCH 1/1 alsa: lx6464es - driver for the digigram lx6464es interface
prototype of a driver for the digigram lx6464es 64 channel ethersound interface.
--- include/linux/pci_ids.h | 5 + sound/pci/Kconfig | 10 + sound/pci/Makefile | 1 + sound/pci/lx6464es/LXES_registers.h | 122 +++ sound/pci/lx6464es/Makefile | 13 + sound/pci/lx6464es/PcxErr_e.h | 680 ++++++++++++++++ sound/pci/lx6464es/ethersound.h | 55 ++ sound/pci/lx6464es/if_drv_mb.h | 287 +++++++ sound/pci/lx6464es/lx6464es.c | 1179 ++++++++++++++++++++++++++++ sound/pci/lx6464es/lx6464es.h | 119 +++ sound/pci/lx6464es/lx_core.c | 1453 +++++++++++++++++++++++++++++++++++ sound/pci/lx6464es/lx_core.h | 239 ++++++ 12 files changed, 4163 insertions(+), 0 deletions(-) create mode 100644 sound/pci/lx6464es/LXES_registers.h create mode 100644 sound/pci/lx6464es/Makefile create mode 100644 sound/pci/lx6464es/PcxErr_e.h create mode 100644 sound/pci/lx6464es/ethersound.h create mode 100644 sound/pci/lx6464es/if_drv_mb.h create mode 100644 sound/pci/lx6464es/lx6464es.c create mode 100644 sound/pci/lx6464es/lx6464es.h create mode 100644 sound/pci/lx6464es/lx_core.c create mode 100644 sound/pci/lx6464es/lx_core.h
include/linux/pci_ids.h +5 -0
@@ -976,6 +976,7 @@#define PCI_DEVICE_ID_PLX_PCI200SYN 0x3196#define PCI_DEVICE_ID_PLX_9030 0x9030#define PCI_DEVICE_ID_PLX_9050 0x9050 +#define PCI_DEVICE_ID_PLX_9056 0x9056#define PCI_DEVICE_ID_PLX_9080 0x9080#define PCI_DEVICE_ID_PLX_GTEK_SERIAL2 0xa001
@@ -1813,6 +1814,10 @@#define PCI_SUBDEVICE_ID_HYPERCOPE_METRO 0x0107#define PCI_SUBDEVICE_ID_HYPERCOPE_CHAMP2 0x0108 +#define PCI_VENDOR_ID_DIGIGRAM 0x1369 +#define PCI_SUBDEVICE_ID_DIGIGRAM_LX6464ES_SERIAL_SUBSYSTEM 0xc001 +#define PCI_SUBDEVICE_ID_DIGIGRAM_LX6464ES_CAE_SERIAL_SUBSYSTEM 0xc002 +#define PCI_VENDOR_ID_KAWASAKI 0x136b#define PCI_DEVICE_ID_MCHIP_KL5A72002 0xff01
sound/pci/Kconfig +10 -0
@@ -622,6 +622,16 @@ config SND_KORG1212 To compile this driver as a module, choose M here: the module will be called snd-korg1212. +config SND_LX6464ES + tristate "Digigram LX6464ES" + select SND_PCM + help + Say Y here to include support for Digigram LX6464ES boards. + + To compile this driver as a module, choose M here: the module + will be called snd-lx6464es. + + config SND_MAESTRO3 tristate "ESS Allegro/Maestro3" select SND_AC97_CODEC
sound/pci/Makefile +1 -0
@@ -62,6 +62,7 @@ obj-$(CONFIG_SND) += \ ca0106/ \ cs46xx/ \ cs5535audio/ \ + lx6464es/ \ echoaudio/ \ emu10k1/ \ hda/ \
sound/pci/lx6464es/LXES_registers.h new file +122 -0
122 lines not on this page · raw message
sound/pci/lx6464es/Makefile new file +13 -0
@@ -0,0 +1,13 @@ +ifeq ($(KERNELRELEASE),) +snd-lx6464es-objs := lx6464es.o lx_core.o +obj-m += snd-lx6464es.o +all: + make -C /lib/modules/$(shell uname -r)/build M=$(PWD) modules + +clean: + make -C /lib/modules/$(shell uname -r)/build M=$(PWD) clean + +else +snd-lx6464es-objs := lx6464es.o lx_core.o +obj-$(CONFIG_SND_LX6464ES) += snd-lx6464es.o +endif
sound/pci/lx6464es/PcxErr_e.h new file +680 -0
680 lines not on this page · raw message
sound/pci/lx6464es/ethersound.h new file +55 -0
55 lines not on this page · raw message
sound/pci/lx6464es/if_drv_mb.h new file +287 -0
287 lines not on this page · raw message
sound/pci/lx6464es/lx6464es.c new file +1179 -0
1,179 lines not on this page · raw message
sound/pci/lx6464es/lx6464es.h new file +119 -0
119 lines not on this page · raw message
sound/pci/lx6464es/lx_core.c new file +1453 -0
505 lines not on this page · raw message
sound/pci/lx6464es/lx_core.h new file +239 -0
239 lines not on this page · raw message
The rest of this diff is not shown. View the raw message.
Signature · 5 lines
1.6.2.1 _______________________________________________ Alsa-devel mailing list [email protected] http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
Part of the diff is not shown here. View the raw message.
Re: [PATCH] alsa: lx6464es - driver for the digigram lx6464es interface
At Mon, 23 Mar 2009 11:12:22 +0100,
Quoted 5 lines · Tim Blechmann
prototype of a driver for the digigram lx6464es 64 channel ethersound interface.
Signed-off-by Tim Blechmann <[email protected]>
Thanks.
First off, the current code looks a bit different from the standard kernel coding style. Please run $LINUX/scripts/checkpatch.pl and fix the issues reported there.
sound/pci/lx6464es/LXES_registers.h line 1
Quoted 7 lines
sound/pci/lx6464es/LXES_registers.h new file +1
-0@@ -0,0 +1,122 @@ +/**
The comments with "/**" in the kernel codes are used for kernel-doc in general. Unless you want to process it via kernel-doc, avoid it.
sound/pci/lx6464es/LXES_registers.h line 2
Quoted 1 line
sound/pci/lx6464es/LXES_registers.h new file +1
-0+* Copyright (C) 2006 DIGIGRAM S.A.
Hrm, is it the copy from Digigram's source? Then we'd need their sign-off as well...
sound/pci/lx6464es/LXES_registers.h line 122
Quoted 3 lines
sound/pci/lx6464es/LXES_registers.h new file +3
-0+//********************************************************************************* +// $history:$ +//
Better to remove unneeded parts like this.
Quoted 9 lines
sound/pci/lx6464es/Makefile new file +3
-0@@ -0,0 +1,13 @@ +ifeq ($(KERNELRELEASE),) +snd-lx6464es-objs := lx6464es.o lx_core.o +obj-m += snd-lx6464es.o
sound/pci/lx6464es/Makefile line 13
Quoted 10 lines
sound/pci/lx6464es/Makefile new file +10
-0+all: + make -C /lib/modules/$(shell uname -r)/build M=$(PWD) modules + +clean: + make -C /lib/modules/$(shell uname -r)/build M=$(PWD) clean + +else +snd-lx6464es-objs := lx6464es.o lx_core.o +obj-$(CONFIG_SND_LX6464ES) += snd-lx6464es.o +endif
Don't put it the stuff for custom builds.
sound/pci/lx6464es/PcxErr_e.h line 42
Quoted 9 lines
sound/pci/lx6464es/PcxErr_e.h new file +4
-0+// Return value when OK +// ******************** + +#define SUCCESS 0
Isn't it defined anywhere?
If it's an internal value, I'd suggest to put a prefix to avoid confusion. No big issue, though.
Quoted 5 lines
sound/pci/lx6464es/lx6464es.c new file +0
-0
...
sound/pci/lx6464es/lx6464es.c line 59
Quoted 3 lines
sound/pci/lx6464es/lx6464es.c new file +3
-0+#ifndef PCI_DEVICE_ID_PLX_9056 /* LATER: this should go to pci_ids.h */ +#define PCI_DEVICE_ID_PLX_9056 0x9056 +#endif
You patch already includes the change in pci_ids.h :) Remove this.
sound/pci/lx6464es/lx6464es.c line 70
Quoted 10 lines
sound/pci/lx6464es/lx6464es.c new file +10
-0+#define PCI_DEVICE_ID_PLX_LX6464ES PCI_DEVICE_ID_PLX_9056 + +#ifndef PCI_VENDOR_ID_DIGIGRAM +#define PCI_VENDOR_ID_DIGIGRAM 0x1369 +#endif + +#ifndef PCI_SUBDEVICE_ID_DIGIGRAM_LX6464ES_SERIAL_SUBSYSTEM +#define PCI_SUBDEVICE_ID_DIGIGRAM_LX6464ES_SERIAL_SUBSYSTEM 0xc001 +#define PCI_SUBDEVICE_ID_DIGIGRAM_LX6464ES_CAE_SERIAL_SUBSYSTEM 0xc002 +#endif
Ditto.
Quoted 1 line
+static int
lx_proc_create(struct snd_card *card, struct lx6464es *chip)
Can be __devinit.
Quoted 3 lines
sound/pci/lx6464es/lx6464es.c new file +3
-0+static int __devinit snd_lx6464es_create (struct snd_card *card, +struct pci_dev *pci, +struct lx6464es **rchip)
...
sound/pci/lx6464es/lx6464es.c line 1046
Quoted 6 lines
sound/pci/lx6464es/lx6464es.c new file +6
-0+ chip->port_mem =pci_resource_start (pci,0 ); + +/* plx port */ + chip->port_plx =pci_resource_start (pci,1 ); + chip->port_plx_remapped =ioport_map (chip->port_plx, +pci_resource_len (pci,1 ));
Better to use pci_ioremap_bar().
sound/pci/lx6464es/lx6464es.c line 1052
Quoted 5 lines
sound/pci/lx6464es/lx6464es.c new file +5
-0+/* dsp port */ + chip->port_dsp =pci_resource_start (pci,2 ); + chip->port_dsp_resource =request_mem_region (chip->port_dsp, +pci_resource_len (pci,2 ), + card_name);
Hm, isn't it already requested via pci_request_regions()?
sound/pci/lx6464es/lx6464es.c line 1054
Quoted 2 lines
sound/pci/lx6464es/lx6464es.c new file +2
-0+ chip->port_dsp_remapped =ioremap_nocache (chip->port_dsp, +pci_resource_len (pci,2 ));
Use pci_ioremap_bar().
sound/pci/lx6464es/lx6464es.c line 1066
Quoted 11 lines
sound/pci/lx6464es/lx6464es.c new file +11
-0+ err =request_irq (pci->irq, lx_interrupt,IRQF_SHARED , + card_name, chip); +if (err) { + snd_printk(KERN_ERR LXP "unable to grab IRQ %d\n", pci->irq); + goto request_irq_failed; + } + chip->irq = pci->irq; + + err =snd_device_new (card,SNDRV_DEV_LOWLEVEL , chip, &ops); +if (err <0 ) +return err;
This error path can be a leak.
sound/pci/lx6464es/lx_core.h line 236
Quoted 13 lines
sound/pci/lx6464es/lx_core.h new file +8
-0+static inline void unpack_pointer (dma_addr_t ptr,u32 *r_low,u32 *r_high) +{ + *r_low = (u32)(ptr &0xffffffff ); +#if BITS_PER_LONG ==32 + *r_high =0 ; +#else + *r_high = (u32)((u64 )ptr>>32 ); +#endif
Better to use upper_32_bits() macro.
thanks,
Takashi
Re: [PATCH] alsa: lx6464es - driver for the digigram lx6464es interface
hi takashi,
i tried to address the issues ...
Tim Blechmann (1):
alsa: lx6464es - driver for the digigram lx6464es interface
include/linux/pci_ids.h | 5 + sound/pci/Kconfig | 10 + sound/pci/Makefile | 1 + sound/pci/lx6464es/Makefile | 2 + sound/pci/lx6464es/lx6464es.c | 1152 ++++++++++++++++++++++++++++++++ sound/pci/lx6464es/lx6464es.h | 114 ++++ sound/pci/lx6464es/lx_core.c | 1442 +++++++++++++++++++++++++++++++++++++++++ sound/pci/lx6464es/lx_core.h | 242 +++++++ sound/pci/lx6464es/lx_defs.h | 376 +++++++++++ 9 files changed, 3344 insertions(+), 0 deletions(-) create mode 100644 sound/pci/lx6464es/Makefile create mode 100644 sound/pci/lx6464es/lx6464es.c create mode 100644 sound/pci/lx6464es/lx6464es.h create mode 100644 sound/pci/lx6464es/lx_core.c create mode 100644 sound/pci/lx6464es/lx_core.h create mode 100644 sound/pci/lx6464es/lx_defs.h
_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
PATCH v2 1/1 alsa: lx6464es - driver for the digigram lx6464es interface
In reply to Takashi Iwaiprototype of a driver for the digigram lx6464es 64 channel ethersound interface.
--- include/linux/pci_ids.h | 5 + sound/pci/Kconfig | 10 + sound/pci/Makefile | 1 + sound/pci/lx6464es/Makefile | 2 + sound/pci/lx6464es/lx6464es.c | 1152 ++++++++++++++++++++++++++++++++ sound/pci/lx6464es/lx6464es.h | 114 ++++ sound/pci/lx6464es/lx_core.c | 1442 +++++++++++++++++++++++++++++++++++++++++ sound/pci/lx6464es/lx_core.h | 242 +++++++ sound/pci/lx6464es/lx_defs.h | 376 +++++++++++ 9 files changed, 3344 insertions(+), 0 deletions(-) create mode 100644 sound/pci/lx6464es/Makefile create mode 100644 sound/pci/lx6464es/lx6464es.c create mode 100644 sound/pci/lx6464es/lx6464es.h create mode 100644 sound/pci/lx6464es/lx_core.c create mode 100644 sound/pci/lx6464es/lx_core.h create mode 100644 sound/pci/lx6464es/lx_defs.h
include/linux/pci_ids.h +5 -0
@@ -976,6 +976,7 @@#define PCI_DEVICE_ID_PLX_PCI200SYN 0x3196#define PCI_DEVICE_ID_PLX_9030 0x9030#define PCI_DEVICE_ID_PLX_9050 0x9050 +#define PCI_DEVICE_ID_PLX_9056 0x9056#define PCI_DEVICE_ID_PLX_9080 0x9080#define PCI_DEVICE_ID_PLX_GTEK_SERIAL2 0xa001
@@ -1813,6 +1814,10 @@#define PCI_SUBDEVICE_ID_HYPERCOPE_METRO 0x0107#define PCI_SUBDEVICE_ID_HYPERCOPE_CHAMP2 0x0108 +#define PCI_VENDOR_ID_DIGIGRAM 0x1369 +#define PCI_SUBDEVICE_ID_DIGIGRAM_LX6464ES_SERIAL_SUBSYSTEM 0xc001 +#define PCI_SUBDEVICE_ID_DIGIGRAM_LX6464ES_CAE_SERIAL_SUBSYSTEM 0xc002 +#define PCI_VENDOR_ID_KAWASAKI 0x136b#define PCI_DEVICE_ID_MCHIP_KL5A72002 0xff01
sound/pci/Kconfig +10 -0
@@ -622,6 +622,16 @@ config SND_KORG1212 To compile this driver as a module, choose M here: the module will be called snd-korg1212. +config SND_LX6464ES + tristate "Digigram LX6464ES" + select SND_PCM + help + Say Y here to include support for Digigram LX6464ES boards. + + To compile this driver as a module, choose M here: the module + will be called snd-lx6464es. + + config SND_MAESTRO3 tristate "ESS Allegro/Maestro3" select SND_AC97_CODEC
sound/pci/Makefile +1 -0
@@ -62,6 +62,7 @@ obj-$(CONFIG_SND) += \ ca0106/ \ cs46xx/ \ cs5535audio/ \ + lx6464es/ \ echoaudio/ \ emu10k1/ \ hda/ \
sound/pci/lx6464es/Makefile new file +2 -0
@@ -0,0 +1,2 @@ +snd-lx6464es-objs := lx6464es.o lx_core.o +obj-$(CONFIG_SND_LX6464ES) += snd-lx6464es.o
sound/pci/lx6464es/lx6464es.c new file +1152 -0
1,152 lines not on this page · raw message
sound/pci/lx6464es/lx6464es.h new file +114 -0
114 lines not on this page · raw message
sound/pci/lx6464es/lx_core.c new file +1442 -0
1,442 lines not on this page · raw message
sound/pci/lx6464es/lx_core.h new file +242 -0
242 lines not on this page · raw message
sound/pci/lx6464es/lx_defs.h new file +376 -0
8 lines not on this page · raw message
The rest of this diff is not shown. View the raw message.
Signature · 1 line
1.6.2.1
_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
Part of the diff is not shown here. View the raw message.
Re: [PATCH v2] alsa: lx6464es - driver for the digigram lx6464es interface
At Tue, 24 Mar 2009 12:24:35 +0100,
Quoted 5 lines · Tim Blechmann
prototype of a driver for the digigram lx6464es 64 channel ethersound interface.
Signed-off-by Tim Blechmann <[email protected]>
In a quick glance, it looks much better now.
I'll merge it later after more checks (as I'm on vacation now).
Note that the merge to the linux kernel tree will be postponed to 2.6.31 since the merge window for 2.6.30 already started.
Thanks!
Takashi
Re: [PATCH v2] alsa: lx6464es - driver for the digigram lx6464es interface
Quoted 10 lines
Quoted 4 lines
prototype of a driver for the digigram lx6464es 64 channel ethersound interface.
Signed-off-by Tim Blechmann <[email protected]>In a quick glance, it looks much better now.
I'll merge it later after more checks (as I'm on vacation now).Note that the merge to the linux kernel tree will be postponed to 2.6.31 since the merge window for 2.6.30 already started.
thanks, no hurry, i will be able to do more tests during the next few weeks, when i have access to a multichannel setup ...
cheers, tim
Signature · 5 lines
[email protected] http://tim.klingt.org The composer makes plans, music laughs. Morton Feldman
_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
Re: [PATCH v2] alsa: lx6464es - driver for the digigram lx6464es interface
At Wed, 25 Mar 2009 14:56:46 +0100,
Quoted 6 lines · Tim Blechmann
Quoted 6 lines
Quoted 5 lines
prototype of a driver for the digigram lx6464es 64 channel ethersound interface.
Signed-off-by Tim Blechmann <[email protected]>
Quoted 8 lines
Quoted 5 lines
In a quick glance, it looks much better now.
I'll merge it later after more checks (as I'm on vacation now).Note that the merge to the linux kernel tree will be postponed to 2.6.31 since the merge window for 2.6.30 already started.
thanks, no hurry, i will be able to do more tests during the next few weeks, when i have access to a multichannel setup ...
Now I merged the driver into sound git tree. I fixed the build of alsa-driver-build tree with that merge (and others), so the latest alsa-driver snapshot should work as is.
Please let me know any issues you find.
thanks,
Takashi
Quoted 12 lines
cheers, tim
Signature · 7 lines
[email protected] http://tim.klingt.org The composer makes plans, music laughs. Morton Feldman [2 OpenPGP digital signature <application/pgp-signature (7bit)>]
Quoted 6 lines
Quoted 2 lines
takashi, i would like to submit the patch to alsa soon ... how should i proceed? can you do a review of the code?
Simply post patches to alsa-devel ML and Cc to me (or vice versa). But I'll be on vacation from this week (Wednesday) until the Easter holidays. So, sooner better ;)
the patch seems to be too large for the mailing-list, 140kb ... i hope
the moderator will accept the message ...
please note, that i included 4 header files and some minor code portions from digigram's windows driver, which do not match the linux coding style. i preferred to do so in order to keep the similarity to the windows driver, which was my main reference for implementing the driver.
cheers, tim
Signature · 5 lines
[email protected] http://tim.klingt.org The composer makes plans, music laughs. Morton Feldman
_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
At Mon, 23 Mar 2009 11:26:11 +0100,
Quoted 10 lines · Tim Blechmann
Quoted 7 lines
Quoted 3 lines
takashi, i would like to submit the patch to alsa soon ... how should i proceed? can you do a review of the code?
Simply post patches to alsa-devel ML and Cc to me (or vice versa). But I'll be on vacation from this week (Wednesday) until the Easter holidays. So, sooner better ;)
the patch seems to be too large for the mailing-list, 140kb ... i hope
the moderator will accept the message ...
Ah, that's the problem, yes. Will check later.
Quoted 4 lines
please note, that i included 4 header files and some minor code portions from digigram's windows driver, which do not match the linux coding style. i preferred to do so in order to keep the similarity to the windows driver, which was my main reference for implementing the driver.
Well, basically it doesn't help to keep the non-standard coding style. Once after your code is merged to the kernel tree, it's no more your baby. Everyone will modify the code, and one of people's favorite actions is to re-format codes to improve the readability :)
So, if it's intended to be merged, please follow the standard coding style.
thanks,
Takashi
Quoted 9 lines
Quoted 4 lines
please note, that i included 4 header files and some minor code portions from digigram's windows driver, which do not match the linux coding style. i preferred to do so in order to keep the similarity to the windows driver, which was my main reference for implementing the driver.
Well, basically it doesn't help to keep the non-standard coding style. Once after your code is merged to the kernel tree, it's no more your baby. Everyone will modify the code, and one of people's favorite actions is to re-format codes to improve the readability :)
ok, i'll write a new header file for the specific definitions ... that's probably better than messing with Lindent or similar to reformat the digigram sources ...
and thanks for the other comments ... i'll adapt the code ...
cheers, tim
Signature · 5 lines
[email protected] http://tim.klingt.org I don't write music for sissy ears. Charles Ives
_______________________________________________
Alsa-devel mailing list
[email protected]
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel