aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/ieee1394
Commit message (Collapse)AuthorAge
* Make ieee1394_init a fs-initcallLinus Torvalds2009-02-26
| | | | | | | | | | | | | | | | | It needs to happen before any firewire driver actually registers itself, and that was previously handled by having the Makefile list the core ieee1394 files before the drivers. But now there are firewire drivers in drivers/media, and the Makefile games aren't enough. So just make ieee1394_init happen earlier in the init sequence, the way all other bus layers already do. Reported-and-tested-by: Ingo Molnar <mingo@elte.hu> Cc: Stefan Richter <stefanr@s5r6.in-berlin.de> Cc: Henrik Kurelid <henrik@kurelid.se> Cc: Mauro Carvalho Chehab <mchehab@infradead.org> Cc: Ben Backx <ben@bbackx.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* ieee1394: remove superfluous assertionsStefan Richter2009-02-24
| | | | | | | | hpsb_read, hpsb_write, hpsb_lock are sleeping functions which nobody is in danger to use in atomic context. Besides, in_interrupt does not cover all types of atomic context. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* ieee1394: inherit ud vendor_id from node vendor_idStefan Richter2009-02-24
| | | | | | | | | | | | | | | | | While Module_Vendor_ID in the configuration ROM's root directory is mandatory, there often aren't vendor IDs in unit directories. This affects the new firedtv driver which is meant to be auto-loaded and matched only for vendor-specific devices. We now always copy ne->vendor_id into ud->vendor_id before we scan a unit directory (and fill in a possibly present vendor ID from there). This way, the root directory's vendor ID is used as fallback in the "uevent" environment for modprobe'ing per module alias when a node was plugged in, and in the driver match routine when protocol drivers are bound to unit directories. It will however not be used as sysfs attribute of a unit directory device. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* ieee1394: add hpsb_node_read() and hpsb_node_lock()Stefan Richter2009-02-24
| | | | | | | | | | | | These will be used by the firedtv driver. Like hpsb_node_write() they are much better APIs for high-level drivers than hpsb_write() and its siblings --- easier to use correctly and also terser. Unlike hspb_node_write(), the two new functions will only be used by one call site. Hence make them static inline instead of exported symbols. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* ieee1394: use correct barrier types between accesses of nodeid and generationStefan Richter2009-02-24
| | | | | | | | | | | | | | A compiler barrier (explicit on the read side, implicit on the write side) is not quite enough for what has to be accomplished here. Use hardware memory barriers on systems which need them. (Of course a full fix of generation handling would require much more than this. The ieee1394 core's bus generation counter had to be tied to the controller's bus generation counter; cf. Kristian's stack. It's just that I have other current business with the code around these barrier()s, so why not do at least this small fix.) Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* firesat: copyrights, rename to firedtv, API conversions, fix remote control ↵Stefan Richter2009-02-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | input Combination of the following changes: Tue, 26 Aug 2008 00:17:30 +0200 (CEST) firedtv: fix remote control input and update the scancode-to-keycode mapping to a current model. Per default, various media key keycodes are emitted which closely match what is printed on the remote. Userland can modify the mapping by means of evdev ioctls. (Not tested.) The old scancode-to-keycode mapping is left in the driver but cannot be modified by ioctls. This preserves status quo for old remotes. Tue, 26 Aug 2008 00:11:28 +0200 (CEST) firedtv: replace tasklet by workqueue job Non-atomic context is a lot nicer to work with. Sun, 24 Aug 2008 23:30:00 +0200 (CEST) firedtv: move some code back to ieee1394 core Partially reverts "ieee1394: remove unused code" of Linux 2.6.25. Sun, 24 Aug 2008 23:29:30 +0200 (CEST) firedtv: replace semaphore by mutex firesat->avc_sem and ->demux_sem have been used exactly like a mutex. The only exception is the schedule_remotecontrol tasklet which did a down_trylock in atomic context. This is not possible with mutex_trylock; however the whole remote control related code is non-functional anyway at the moment. This should be fixed eventually, probably by turning the tasklet into a worqueue job. Convert everything else from semaphore to mutex. Also rewrite a few of the affected functions to unlock the mutex at a single exit point, instead of in several branches. Sun, 24 Aug 2008 23:28:45 +0200 (CEST) firedtv: some header cleanups Unify #ifndef/#define/#endif guards against multiple inclusion. Drop extern keyword from function declarations. Remove #include's into header files where struct declarations suffice. Remove unused ohci1394 interface and related unused ieee1394 interfaces. Add a few missing #include's and remove a few apparently obsolete ones. Sort them alphabetically. Sun, 24 Aug 2008 23:27:45 +0200 (CEST) firedtv: nicer registration message and some initialization fixes Print the correct name in dvb_register_adapter(). While we are at it, replace two switch cascades by one for loop, remove a superfluous member of struct firesat and of two unused arguments of AVCIdentifySubunit(), and fix bogus kfree's in firesat_dvbdev_init(). Tue, 26 Aug 2008 14:24:17 +0200 (CEST) firesat: rename to firedtv Suggested by Andreas Monitzer. Besides DVB-S/-S2 receivers, the driver also supports DVB-C and DVB-T receivers, hence the previous project name is too narrow now. Not yet done: Rename source directory, files, types, variables... Sun, 24 Aug 2008 23:26:23 +0200 (CEST) firesat: add missing copyright notes Reported by Andreas Monitzer and Christian Dolzer. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* Merge branch 'for-linus' of ↵Linus Torvalds2009-02-06
|\ | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6 * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6: ieee1394: dv1394: move deprecation message from module init to file open firewire: core: Remove card from list of cards when enable fails
| * ieee1394: dv1394: move deprecation message from module init to file openStefan Richter2009-02-06
| | | | | | | | | | | | | | | | | | | | | | | | On many Linux installations, the dv1394 driver will be auto-loaded whenever an AV/C device (e.g. camcorder or audio device) is plugged in. An irritating message would then appear in the kernel log. Defer this message to until a dv1394 character device file is actually used by a program. Also include the program name in the message and update the message slightly. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* | Merge branch 'for-linus' of ↵Linus Torvalds2009-01-29
|\| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6 * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6: ieee1394: sbp2: add workarounds for 2nd and 3rd generation iPods firewire: sbp2: add workarounds for 2nd and 3rd generation iPods firewire: sbp2: fix DMA mapping leak on the failure path firewire: sbp2: define some magic numbers as macros firewire: sbp2: fix payload limit at S1600 and S3200 ieee1394: sbp2: don't assume zero model_id or firmware_revision if there is none ieee1394: sbp2: fix payload limit at S1600 and S3200 ieee1394: sbp2: update a help string ieee1394: support for speeds greater than S800 firewire: core: optimize card shutdown ieee1394: ohci1394: increase AT req. retries, fix ack_busy_X from Panasonic camcorders and others firewire: ohci: increase AT req. retries, fix ack_busy_X from Panasonic camcorders and others firewire: ohci: change "context_stop: still active" log message firewire: keep highlevel drivers attached during brief connection loss firewire: unnecessary BM delay after generation rollover firewire: insist on successive self ID complete events
| * ieee1394: sbp2: add workarounds for 2nd and 3rd generation iPodsStefan Richter2009-01-29
| | | | | | | | | | | | | | | | | | | | as per https://bugs.launchpad.net/bugs/294391. These got one sample of each iPod generation going. However there still occurred I/O stalls with the 3rd generation iPod which remain undiagnosed at the time of this writing. Acked-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: sbp2: don't assume zero model_id or firmware_revision if there is noneStefan Richter2009-01-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This makes sbp2 behave more like firewire-sbp2 which reports 0xff000000 as immediate value if there are no unit directory entries for model_id or firmware_revision. It does not reduce matches with the currently existing quirks table; the only zero entry there is for a device which actually does have a zero model_id. It only changes how model_id and firmware_revision are logged if they are missing. Other functionally unrelated changes: The model_id member of quirks list entries is renamed to model; the value (but not the effect) of SBP2_ROM_VALUE_WILDCARD is changed. Now this part of the source is identical with firewire-sbp2 for easier maintenance. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: sbp2: fix payload limit at S1600 and S3200Stefan Richter2009-01-28
| | | | | | | | | | | | | | | | | | | | 1394-2008 clause 16.3.4.1 (1394b-2002 clause 16.3.1.1) defines tighter limits than 1394-2008 clause 6.2.2.3 (1394a-2000 clause 6.2.2.3). Our previously too large limit doesn't matter though if the controller reports its max_receive correctly. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: sbp2: update a help stringStefan Richter2009-01-28
| | | | | | | | Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: support for speeds greater than S800Stefan Richter2009-01-28
| | | | | | | | | | | | | | | | | | | | | | | | The hard-wired configuration of the top speed (until now S800) was unnecessary, remove it. If the local link layer controller supports S1600 or S3200, we now assume this speed for all present 1394b PHYs (except if they are behind 1394a repeaters) until nodemgr figured out the actual speed while fetching the config ROM. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: ohci1394: increase AT req. retries, fix ack_busy_X from Panasonic ↵Stefan Richter2009-01-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | camcorders and others Camcorders have a tendency to fail read requests to their config ROM and write request to their FCP command register with ack_busy_X. This has become a problem with newer kernels and especially Panasonic camcorders, causing AV/C in dvgrab and kino to fail. Dvgrab for example frequently logs "send oops"; kino reports loss of AV/C control. I suspect that lower CPU scheduling latencies in newer kernels made this issue more prominent now. According to https://sourceforge.net/tracker/?func=detail&atid=114103&aid=2492640&group_id=14103 this can be fixed by configuring the FireWire controller for more hardware retries for request transmission; these retries are evidently more successful than libavc1394's own retry loop (typically 3 tries on top of hardware retries). Presumably the same issue has been reported at https://bugzilla.redhat.com/show_bug.cgi?id=449252 and https://bugzilla.redhat.com/show_bug.cgi?id=477279 . Tested-by: Mathias Beilstein Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* | i2c: Delete many unused adapter IDsJean Delvare2009-01-26
|/ | | | Signed-off-by: Jean Delvare <khali@linux-fr.org>
* Merge branch 'master' of ↵David S. Miller2009-01-08
|\ | | | | | | master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6
| * ieee1934: dv1394: interrupt enabling/disabling broken on big-endianHarvey Harrison2009-01-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After annotating the frame structs, this was left: drivers/ieee1394/dv1394.c:2113:23: warning: invalid assignment: |= drivers/ieee1394/dv1394.c:2113:23: left side has type restricted __le32 drivers/ieee1394/dv1394.c:2113:23: right side has type int drivers/ieee1394/dv1394.c:2121:24: warning: invalid assignment: &= drivers/ieee1394/dv1394.c:2121:24: left side has type restricted __le32 drivers/ieee1394/dv1394.c:2121:24: right side has type int drivers/ieee1394/dv1394.c:2123:24: warning: invalid assignment: |= drivers/ieee1394/dv1394.c:2123:24: left side has type restricted __le32 drivers/ieee1394/dv1394.c:2123:24: right side has type int Which looks like a real bug on a big-endian arch as it would set/clear the wrong bit. Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Bill Fink writes: I finally got a chance to test the patch on my kernel, and live DV viewing using xine still worked fine. Although I admit to being mystified how it works both before and after the patch, since the cpu_to_le32() calls that were added should result in byte swapping on PPC that wasn't being done before. I guess that either the code paths involved aren't actually being triggered by my xine DV viewing, or there's some fortuitous palindromic setting of bits. Tested-by: Bill Fink <billfink@mindspring.com> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: dv1394: annotate frame input/output structs as little endianHarvey Harrison2009-01-04
| | | | | | | | | | | | | | No Functional changes. Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: eth1394: trivial sparse annotationsHarvey Harrison2009-01-04
| | | | | | | | | | | | | | Mostly annotations of ether_type as a be16. Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: mark bus_info_data as a __be32 arrayHarvey Harrison2009-01-04
| | | | | | | | | | | | | | | | | | | | Two access functions get_max_rom and set_hw_config_rom are changed to take __be32 as well. Only bus_info_data was ever passed in so this is OK. All other uses of bus_info_data treated it as a be32 value already. Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: replace CSR_SET_BUS_INFO_GENERATION macroHarvey Harrison2009-01-04
| | | | | | | | | | Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: pcilynx: trivial endian annotationHarvey Harrison2009-01-04
| | | | | | | | | | | | | | | | bus_info_block was treated as a be32 everywhere, annotate as such. Removes plenty of sparse warnings. Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: ignore nonzero Bus_Info_Block.max_rom, fetch config ROM in quadletsStefan Richter2009-01-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It is already known that buggy firmwares exist which report a bogus link_spd in their config ROM bus info block. We now got the first report of a bogus max_rom too (Freecom FireWire Hard Drive 1TB, http://bugzilla.kernel.org/show_bug.cgi?id=12206). I suspect other OSs only use quadlet reads to fetch the config ROM, otherwise the firmware authors would have noticed their mistake. Hence limit ieee1394's config ROM fetching routine to quadlets as the safe minimum regardless of what the bus info block says. This will potentially slow the bus reset handling by nodemgr somewhat down. But most existing devices support only quadlet reads anyway, hence there will often be no actual difference to before this change. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: consolidate uses of IEEE1934_BUSID_MAGICHarvey Harrison2009-01-04
| | | | | | | | | | | | | | Move the definition out of nodemgr.h and use it in csr.c/pcilynx.c Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: ohci1394: flush MMIO writes before delay in initializationStefan Richter2009-01-04
| | | | | | | | | | | | and replace busy-wait by msleep. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: ohci1394: pass error codes from request_irq throughStefan Richter2009-01-04
| | | | | | | | Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: ohci1394: don't leave interrupts enabled during suspend/resumeFrans Pop2009-01-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On my HP 2510p I get the following in dmesg during near the end of most resumes from suspend to RAM: irq 19: nobody cared (try booting with the "irqpoll" option) Pid: 0, comm: swapper Not tainted 2.6.28-rc7 #67 Call Trace: <IRQ> [<ffffffffa00ee9e1>] ? ohci_irq_handler+0x60/0x7e9 [ohci1394] [<ffffffff8026aa4d>] __report_bad_irq+0x38/0x87 [<ffffffff8026abaa>] note_interrupt+0x10e/0x174 [<ffffffff8026b262>] handle_fasteoi_irq+0xa7/0xd1 [<ffffffff8020eb87>] do_IRQ+0x73/0xe4 [<ffffffff8020c626>] ret_from_intr+0x0/0xa <EOI> [<ffffffffa0012606>] ? acpi_idle_enter_bm+0x26b/0x2b2 [processor] [<ffffffffa00125fc>] ? acpi_idle_enter_bm+0x261/0x2b2 [processor] [<ffffffff8024f30f>] ? notifier_call_chain+0x33/0x5b [<ffffffff803b9c64>] ? cpuidle_idle_call+0x8c/0xc4 [<ffffffff8020b312>] ? cpu_idle+0x4a/0x9a [<ffffffff8042c5c8>] ? rest_init+0x5c/0x5e handlers: [<ffffffffa00ee981>] (ohci_irq_handler+0x0/0x7e9 [ohci1394]) Disabling IRQ #19 There also seems to be an interrupt storm during suspend/resume when this happens: 19: 99968 33 IO-APIC-fasteoi ohci1394 This patch gets rid of both issues and makes the resume as a whole significantly faster. Signed-off-by: Frans Pop <elendil@planet.nl> As was pointed out in http://lkml.org/lkml/2008/12/6/127, this does not fix the cause of the interrupt storm. However, since the source of the interrupts could not be determined yet, we make the system at least more usable with this change. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: mark all hpsb_address_ops instances as constStefan Richter2009-01-04
| | | | | | | | | | | | These are never modified. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: replace a GFP_ATOMIC by GFP_KERNEL allocationStefan Richter2009-01-04
| | | | | | | | | | | | All callers of hpsb_register_addrspace() can sleep. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* | ieee1394: use internal network device statsStephen Hemminger2009-01-06
| | | | | | | | | | | | | | | | Use the network_device_stats field in network_device. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Acked-by: Stefan Richter <stefanr@s5r6.in-berlin.de> Signed-off-by: David S. Miller <davem@davemloft.net>
* | ieee1394: remove unneeded last_rxStephen Hemminger2009-01-06
| | | | | | | | | | | | | | | | Last_rx is now done if needed inside bonding. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Acked-by: Stefan Richter <stefanr@s5r6.in-berlin.de> Signed-off-by: David S. Miller <davem@davemloft.net>
* | ieee1394: convert to net_device_opsStephen Hemminger2009-01-06
|/ | | | | | | | Convert to net_device_ops. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Acked-by: Stefan Richter <stefanr@s5r6.in-berlin.de> Signed-off-by: David S. Miller <davem@davemloft.net>
* ieee1394: add quirk fix for Freecom HDDStefan Richter2008-12-13
| | | | | | | | | | According to http://bugzilla.kernel.org/show_bug.cgi?id=12206, Freecom FireWire Hard Drive 1TB reports max_rom=2 but returns garbage if block read requests are used to read the config ROM. Force max_rom=0 to limit them to quadlet read requests. Reported-by: Christian Mueller <cm1@mumac.de> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* ieee1394: node manager causes up to ~3.25s delay in freezing tasksNigel Cunningham2008-12-09
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The firewire nodemanager function "nodemgr_host_thread" contains a loop that calls try_to_freeze near the top of the loop, but then delays for up to 3.25 seconds (plus time to do work) before getting back to the top of the loop. When starting a cycle post-boot, this doesn't seem to bite, but it is causing a noticeable delay at boot time, when freezing processes prior to starting to read the image. The following patch adds invocation of try_to_freeze to the subloops that are used in the body of this function. With these additions, the time to freeze when starting to resume at boot time is virtually zero. I'm no expert on firewire, and so don't know that we shouldn't check the return value and jump back to the top of the loop or such like after being frozen, but I submit it for your consideration. Signed-off-by: Nigel Cunningham <nigel@tuxonice.net> The delay until nodemgr freezes was up to 0.25s (plus time for node probes) in Linux 2.6.27 and older and up to 3.25s (plus ~) since Linux 2.6.28-rc1, hence much more noticeable. try_to_freeze() without any jump is correct. The surrounding code in the respective loops will catch whether another bus reset happens during the freeze and handle it. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* ieee1394: sbp2: fix race condition in state changeStefan Richter2008-11-29
| | | | | | | An intermediate transition from _RUNNING to _IN_SHUTDOWN could have been missed by the former code. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* ieee1394: fix list corruption (reported at module removal)Stefan Richter2008-11-29
| | | | | | | | | | | | | | If there is more than one FireWire controller present, dummy_zero_addr and dummy_max_addr were added multiple times to different lists, thus corrupting the lists. Fix this by allocating them dynamically per host instead of just once globally. (Perhaps a better address space allocation algorithm could rid us of the two dummy address spaces.) Fixes http://bugzilla.kernel.org/show_bug.cgi?id=10129 . Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* ieee1394: sbp2: another iPod mini quirk entryStefan Richter2008-11-25
| | | | | | | | Add another model ID of a broken firmware to prevent early I/O errors by acesses at the end of the disk. Reported at linux1394-user, http://marc.info/?t=122670842900002 Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* Merge branch 'for-linus' of ↵Linus Torvalds2008-11-06
|\ | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6 * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6: ieee1394: dv1394: fix possible deadlock in multithreaded clients ieee1394: raw1394: fix possible deadlock in multithreaded clients ieee1394: struct device - replace bus_id with dev_name(), dev_set_name() firewire: struct device - replace bus_id with dev_name(), dev_set_name()
| * ieee1394: dv1394: fix possible deadlock in multithreaded clientsStefan Richter2008-10-31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix a possible though highly unlikely deadlock: Thread A: Thread B: - acquire mmap_sem - dv1394_ioctl/read/write() - dv1394_mmap() - acquire video->mtx - acquire video->mtx - copy_to/from_user(), possible page fault: acquire mmap_sem The simplest fix is to use mutex_trylock() instead of mutex_lock() in dv1394_mmap(). This changes the behavior under contention in a way which is visible to userspace clients. However, my guess is that no clients exist which use mmap vs. ioctl/read/write on the dv1394 character device file interface in concurrent threads. Reported-by: Johannes Weiner <hannes@saeurebad.de> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: raw1394: fix possible deadlock in multithreaded clientsStefan Richter2008-10-31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Regression in 2.6.28-rc1: When I added the new state_mutex which prevents corruption of raw1394's internal state when accessed by multithreaded client applications, the following possible though highly unlikely deadlock slipped in: Thread A: Thread B: - acquire mmap_sem - raw1394_write() or raw1394_ioctl() - raw1394_mmap() - acquire state_mutex - acquire state_mutex - copy_to/from_user(), possible page fault: acquire mmap_sem The simplest fix is to use mutex_trylock() instead of mutex_lock() in raw1394_mmap(). This changes the behavior under contention in a way which is visible to userspace clients. However, since multithreaded access was entirely buggy before state_mutex was added and libraw1394's documentation advised application programmers to use a handle only in a single thread, this change in behaviour should not be an issue in practice at all. Since we have to use mutex_trylock() in raw1394_mmap() regardless whether /dev/raw1394 was opened with O_NONBLOCK or not, we now use mutex_trylock() unconditionally everywhere for state_mutex, just to have consistent behavior. Reported-by: Johannes Weiner <hannes@saeurebad.de> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: struct device - replace bus_id with dev_name(), dev_set_name()Kay Sievers2008-10-31
| | | | | | | | | | | | Acked-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
* | saner FASYNC handling on file closeAl Viro2008-11-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | As it is, all instances of ->release() for files that have ->fasync() need to remember to evict file from fasync lists; forgetting that creates a hole and we actually have a bunch that *does* forget. So let's keep our lives simple - let __fput() check FASYNC in file->f_flags and call ->fasync() there if it's been set. And lose that crap in ->release() instances - leaving it there is still valid, but we don't have to bother anymore. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | Merge branch 'for-linus' of ↵Linus Torvalds2008-10-16
|\| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6 * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6: firewire: Add more documentation to firewire-cdev.h firewire: fix ioctl() return code firewire: fix setting tag and sy in iso transmission firewire: fw-sbp2: fix another small generation access bug firewire: fw-sbp2: enforce s/g segment size limit firewire: fw_send_request_sync() ieee1394: survive a few seconds connection loss ieee1394: nodemgr clean up class iterators ieee1394: dv1394, video1394: remove unnecessary expressions ieee1394: raw1394: make write() thread-safe ieee1394: raw1394: narrow down the state_mutex protected region ieee1394: raw1394: replace BKL by local mutex, make ioctl() and mmap() thread-safe ieee1394: sbp2: enforce s/g segment size limit ieee1394: sbp2: check for DMA mapping failures ieee1394: sbp2: stricter dma_sync ieee1394: Use DIV_ROUND_UP
| * ieee1394: survive a few seconds connection lossStefan Richter2008-10-15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are situations when nodes vanish from the bus and come back in quickly thereafter: - When certain bus-powered hubs are plugged in, - when certain disk enclosures are switched from self-power to bus power or vice versa and break the daisy chain during the transition, - when the user plugs a cable out and quickly plugs it back in, e.g. to reorder a daisy chain (works on Mac OS X if done quickly enough), - when certain hubs temporarily malfunction during high bus traffic. The ieee1394 driver's nodemgr already contained a function to set vanished nodes aside into "limbo"; i.e. they wouldn't actually be deleted right away. (In fact, only unloading the driver or writing into an obscure sysfs attribute would delete them eventually.) If nodes reappeared later, they would be resurrected out of limbo. Moving nodes into and out of limbo was accompanied with calling the .suspend() and .resume() driver methods of the drivers which were bound to a respective node's unit directories. Not only is this somewhat strange due to the intended use of these driver methods for power management, also the sbp2 driver in particular does not implement .suspend() and .resume(). Hence sbp2 would be disconnected from devices in situations as listed above. We now: - leave drivers bound when nodes go into limbo, - call the drivers' .update() when nodes come out of limbo, - automatically delete in-limbo nodes 3 seconds after the last bus reset and bus rescan. - Because of the automatic removal, the now obsolete bus attribute /sys/bus/ieee1394/destroy_node is removed. This especially lets sbp2 survive brief disconnections. You can for example yank a disk's cable and plug it back in while reading the respective disk with dd, but dd will happily continue as if nothing happened. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: nodemgr clean up class iteratorsStefan Richter2008-10-15
| | | | | | | | | | | | | | | | | | Remove useless pointer type casts. Remove unnecessary hi->host indirection where only host is used. Remove an unnecessary WARN_ON. Change a few names. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: dv1394, video1394: remove unnecessary expressionsStefan Richter2008-10-15
| | | | | | | | | | | | | | | | init->channel and v.buffer are unsigned and tests for < 0 therefore always false. gcc knows this and eliminates the code, but anyway... Reported by Roel Kluin. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: raw1394: make write() thread-safeStefan Richter2008-10-15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Application programs should use a libraw1394 handle only in a single thread. The raw1394 driver was apparently relying on this, because it did nothing to protect its fi->state variable from corruption due to concurrent accesses. We now serialize the fi->state accesses. This affects the write() path. We re-use the state_mutex which was introduced to protect fi->iso_state accesses in the ioctl() path. These paths and accesses are independent of each other, hence separate mutexes could be used. But I don't see much benefit in that. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: raw1394: narrow down the state_mutex protected regionStefan Richter2008-10-15
| | | | | | | | | | | | | | | | | | | | Refactor the ioctl dispatcher in order to move a fraction of it out of the section which is serialized by fi->state_mutex. This is not so much about performance but more about self-documentation: The mutex_lock()/ mutex_unlock() calls are now closer to the data accesses which the mutex protects, i.e. to the iso_state switch. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
| * ieee1394: raw1394: replace BKL by local mutex, make ioctl() and mmap() ↵Stefan Richter2008-10-15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | thread-safe This removes the last usage of the Big Kernel Lock from the ieee1394 stack, i.e. from raw1394's (unlocked_)ioctl and compat_ioctl. The ioctl()s don't need to take the BKL, but they need to be serialized per struct file *. In particular, accesses to ->iso_state need to be serial. We simply use a blocking mutex for this purpose because libraw1394 does not use O_NONBLOCK. In practice, there is no lock contention anyway because most if not all libraw1394 clients use a libraw1394 handle only in a single thread. mmap() also accesses ->iso_state. Until now this was unprotected against concurrent changes by ioctls. Fix this bug while we are at it. Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>