The Audigy 2 CA0102 chip (but most likely others from the emu10k1 family,
too) has a problem that from time to time it likes to do few DMA reads a
bit beyond its normal allocation and gets very confused if these reads get
blocked by a IOMMU.
For the first (reserved) page this happens multiple times at every
playback, for various synth pages it happens randomly, rarely for PCM
playback buffers and the page table memory itself.
All these reads seem to follow a similar pattern, observed read offsets
beyond the allocation end were 0x00, 0x40, 0x80 and 0xc0 (PCI cache line
multiples), so it looks like the device tries to accesses up to 256 extra
bytes.
As a workaround let's widen these DMA allocations by an extra page if we
detect that the device is behind a non-passthrough IOMMU (the DMA memory
should be relatively plenty on IOMMU systems).
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When we get a IOMMU page fault for a emu10k1 device it is very hard to
discover which of chip many DMA allocations triggered it (since on a IOMMU
system the DMA address space is often very different from the CPU one).
Let's add optional debug printouts providing this information.
These debug printouts are only enabled on an explicit request via the
kernel dynamic debug mechanism.
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
We have been calling dma_set_mask() and then dma_set_coherent_mask() with
the same value, but there is a dma_set_mask_and_coherent() function that
does exactly that so let's use it instead.
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The emu10k1-family chips need the first page (index 0) reserved in their
page tables for some reason (every emu10k1 driver I've checked does this
without much of an explanation).
Using the first page for normal samples results in a broken playback.
However, we already have a dummy page allocated - so called "silent page"
and, in fact, had always been setting it as the first page in the chip page
table because an initialization of every entry of the page table to point
to a silent page happens after and overwrites the reserved_page allocation.
So the only thing remaining to remove the reserved_page allocation is a
trivial change to the page allocation logic to ignore the first page entry
and start its allocations from the second entry (index 1).
Signed-off-by: Maciej S. Szmigiero <mail@maciej.szmigiero.name>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This patch is a cleanup of EMU1010 dock probing code in emu10k1 driver
to use work instead of kthread in a loop. The work is lighter and
easier to control than kthread, in general.
Instead of a loop with the explicit sleep, we do simply
delayed-schedule the work. At suspend/resume callbacks, the work is
canceled and restarted, respectively.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The EMU1010 support in emu10k1 driver has two request_firmware()
calls, one for the main board and one for the dock. Both call
patterns are fairly similar, and we can simplify it by introducing a
helper function and a table instead of the open switch/case.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The emu1010_firmware_thread() checks the previous dock status, but a
wrong register is recorded as the last status when the dock is plugged
in. Usually this isn't a big issue since this value gets overwritten
by the next loop after one second. But when a dock is unplugged
immediately after plugging, it means essentially missing undock
handling.
This patch addresses it by remembering the correct register value.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Since kthread_create can be failed, it needs to check
whether error occurred and return error code.
Signed-off-by: Insu Yun <wuninsu@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Build emuproc.o conditionally and drop the unneeded ifdefs.
Some are replaced with the new CONFIG_SND_PROC_FS.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Looks like audigy emu10k2 (probably emu10k1 - sb live too) support two
modes for DMA. Second mode is useful for 64 bit os with more then 2 GB
of ram (fixes problems with big soundfont loading)
1) 32MB from 2 GB address space using 8192 pages (used now as default)
2) 16MB from 4 GB address space using 4096 pages
Mode is set using HCFG_EXPANDED_MEM flag in HCFG register.
Also format of emu10k2 page table is then different.
Signed-off-by: Peter Zubaj <pzubaj@marticonet.sk>
Tested-by: Takashi Iwai <tiwai@suse.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Some models provide too long string for the shortname that has 32bytes
including the terminator, and it results in a non-terminated string
exposed to the user-space. This isn't too critical, though, as the
string is stopped at the succeeding longname string.
This patch fixes such entries by dropping "SB" prefix (it's enough to
fit within 32 bytes, so far). Meanwhile, it also changes strcpy()
with strlcpy() to make sure that this kind of problem won't happen in
future, too.
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When the dock on an E-mu 1010 card is disconnected, all outputs get
muted by the hardware. Add logic to detect a disconnect and unmute.
Signed-off-by: Michael Gernoth <michael@gernoth.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The Audigy 5/Rx is essentially an Audigy 4 behind a PLX PCIe-
bridge with an additional TOSLINK output.
Signed-off-by: Michael Gernoth <michael@gernoth.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The functions kfree(), release_firmware() and snd_util_memhdr_free() test
whether their argument is NULL and then return immediately. Thus the test
around the call is not needed.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The commit [b209c4df: ALSA: emu10k1: cache emu1010 firmware] broke the
firmware loading of the dock, just (mistakenly) ignoring a different
firmware for docks on some models. This patch revives them again.
Bugzilla: https://bugs.archlinux.org/task/34865
Reported-and-tested-by: Tobias Powalowski <tobias.powalowski@googlemail.com>
Cc: <stable@vger.kernel.org> [v3.8+]
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This expands the regression fix from
d28215996b.
The firmware also needs to be loaded when it was already cached.
Signed-off-by: Florian Zeitz <florob@babelmonkeys.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
CONFIG_HOTPLUG is going away as an option. As result the __dev*
markings will be going away.
Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst,
and __devexit.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Add chip details for E-mu 1010 PCIe card. It has the same
chip as found in E-mu 1010b but it uses different PCI id.
Signed-off-by: Maxim Kachur <mcdebugger@duganet.ru>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Lots of sound drivers were getting module.h via the implicit presence
of it in <linux/device.h> but we are going to clean that up. So
fix up those users now.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
This patch adds the necessary details to support the PCIe version of
E-MU's 0404 card.
From comparing the PCBs it seems the PCIe version just added a PCIe
chipset and left all other components pretty much in place.
For anyone intrigued to take a look at the PCB there are pictures I took
at <http://babelmonkeys.de/~florob/E-MU%200404/>.
Signed-off-by: Florian Zeitz <florob@babelmonkeys.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The name argument of request_irq() appears in /proc/interrupts, and
it's quite ugly when the name entry contains a space or special letters.
In general, it's simpler and more readable when the module name appears
there, so let's replace all entries with KBUILD_MODNAME.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The patch below removes an extra "l" in the word.
Signed-off-by: Justin P. Mattock <justinmattock@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Fix this sparse warnings:
sound/pci/emu10k1/emu10k1_main.c:723:66: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:724:68: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:748:74: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:751:66: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:759:73: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:760:73: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:837:50: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:845:50: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:881:50: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:889:57: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:890:57: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:895:60: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:897:60: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:899:60: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:910:56: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:914:57: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:918:56: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:922:57: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:924:58: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:936:60: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:1073:60: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:1088:60: warning: incorrect type in argument 3 (different signedness)
sound/pci/emu10k1/emu10k1_main.c:1093:58: warning: incorrect type in argument 3 (different signedness)
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
I fixed all of coding style errors and some warnings, now it is down to:
checkpatch.pl-0.24 --no-tree --file --strict --terse emu10k1_main.c
total: 0 errors, 62 warnings, 7 checks, 2075 lines checked
Signed-off-by: Vedran Miletic <rivanvx@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Commit 18c7109289 had #endif leftoff from
compilation. This patch fixes it.
Also, I replaced a misplaced comment by a useful one, that explains why are
here #ifdef and #endif added in compilation.
Signed-off-by: Vedran Miletic <rivanvx@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
* added missing SBxxxx, CTxxxx, PCxxx and MAEMxxxx where they were missing,
and fixed some of them which were wrong (according to kx.inf, which is pretty
accurate compared to anything out there)
* fixed device names to make them more consistent across various cards
* fixed order of devices where appropriate
Signed-off-by: Vedran Miletic <rivanvx@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
On Audigy2 Platinum, the Analog/Digital mixer switch is inverted.
https://bugzilla.novell.com/show_bug.cgi?id=396204
The patch adds a simple workaround.
There might be another device requiring a similar fix, too (or fix for
audigy2 generically), but right now I fix only the known broken one.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When the Linux kernel is compiled with CONFIG_DEBUG_SHIRQ=y,
the Soundblaster Audigy2 ZS Notebook PCMCIA card causes the
system hang during boot (udev stage) or when the card is hot-plug.
The CONFIG_DEBUG_SHIRQ flag is by default 'y' with all Fedora
kernels since 2.6.23. The problem was reported as
https://bugzilla.redhat.com/show_bug.cgi?id=326411
The issue was hunted down to the snd_emu10k1_create() routine:
/* pseudo-code */
snd_emu10k1_create(...) {
...
request_irq(... IRQF_SHARED ...) {
register the irq handler
#ifdef CONFIG_DEBUG_SHIRQ
call the irq handler: snd_emu10k1_interrupt() {
poll I/O port // <---- !! system hangs
...
}
#endif
}
...
snd_emu10k1_cardbus_init(...) {
initialize I/O ports
}
...
}
The early access to I/O port in the interrupt handler causes
the freeze. Obviously it is necessary to init the I/O ports
before accessing them. This patch moves the registration of
the irq handler after the initialization of the I/O ports.
Signed-off-by: Jaroslav Franek <jarin.franek@post.cz>
Acked-by: James Courtier-Dutton <James@superbug.co.uk>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The irq handler of PCI drivers must be released before releasing other
resources since the handler for a shared irq can be still called and
may access the freed resource again.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This is based on pseudo-random playing around with the capabilities.
With ca0102 this card gives no output atall, ca0108 appears to work
fine, so it rather looks similar to the EMU1010b/EMU1010 changes.
Some other people seem to have succeeded in using this aswell:
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3496
From: Veli-Matti Valtonen <maligor@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Don't create emu1010 kthread again at resume if it's already created.
Also make the thread function static.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Kees Cook