It seems unlikely, but access to self_id_cpu[0] could at least in theory
be deferred until after the loop over self_id_cpu[1..n] or even after
the subsequent reg_read. Enforce the desired order by a read barrier.
Also prevent the reg_read from being reordered relative to the for loop.
This isn't necessary if the loop's conditional printk counts as an
implicit barrier, but better make it explicit.
(self_id_cpu[] is a coherent DMA buffer.)
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Firewire-sbp2 did very uncooperative things in the kernel's shared
workqueue: Sleeping until reception of management status from the
target for up to 2 seconds, and performing SCSI inquiry and all of the
setup of SCSI command set drivers via scsi_add_device. If there were
transient or permanent error conditions, this caused long blockage of
the kernel's events process, noticeable e.g. by blocked keyboard input.
We now allocate a workqueue process exclusive to fw-sbp2. As a side
effect, this also increases parallelism of fw-sbp2's login and reconnect
work versus fw-core's device discovery and device update work which is
performed in the shared workqueue.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
On rare occasions, the ability to set one of the workaround flags at
runtime may save the day.
People who experience I/O errors with firewire-sbp2 while the old sbp2
driver worked for them should try workarounds=1 and report to the devel
mailinglist whether that improves things. Firewire-sbp2 defaults to the
SCSI stack's maximum transfer size per command, while sbp2 limits them
to 128 kBytes. Flag 1 accomplishes just that.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
On IOMMU-less noncoherent architectures, orb->callback will memcpy the
whole SCSI command buffer for READ-like SCSI commands. It is therefore
friendlier to enable IRQs before the call, like before patch "Add
ref-counting for sbp2 orbs".
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Acked-by: Kristian Høgsberg <krh@redhat.com>
Sparse warned about it although it was apparently harmless:
drivers/firewire/fw-cdev.c:624:23: warning: symbol 'interrupt' shadows an earlier one
include/asm/hw_irq.h:29:13: originally declared here
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
This changes the uevent buffer functions to use a struct instead of a
long list of parameters. It does no longer require the caller to do the
proper buffer termination and size accounting, which is currently wrong
in some places. It fixes a known bug where parts of the uevent
environment are overwritten because of wrong index calculations.
Many thanks to Mathieu Desnoyers for finding bugs and improving the
error handling.
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Fixes (papers over) "Sleep problems with kernels >= 2.6.21 on powerpc",
http://lkml.org/lkml/2007/8/25/155. The issue is that the FireWire
controller's pci_dev.current_state of iBook G3 and presumably older
PowerBooks is still in PCI_UNKNOWN instead of PCI_D0 when the firewire
driver's .suspend method is called.
Like it was suggested earlier in http://lkml.org/lkml/2006/10/24/13, we
do not fail .suspend anymore if pci_set_power_state failed.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
This handles the case where we get the status write before getting the
complete_transaction callback ("status write for unknown orb"). In
this case, we just assume that the initial orb pointer transaction
succeeded and finish the orb. To prevent the transaction callback
from touching freed memory, we ref-count the orb structures.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Fix panic in run_timer_softirq right after "modprobe -r firewire-ohci"
if a FireWire disk was attached and firewire-sbp2 loaded.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
As far as I know, all CardBus FireWire 400 adapters have a maximum
payload of 1024 bytes which is less than the speed-dependent limit of
2048 bytes. Fw-sbp2 has to take the host adapter's limit into account.
This apparently fixes Juju's incompatibility with my CardBus cards, a
NEC based card and a VIA based card.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Acked-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
collapsed with fw-sbp2 patch "Drop cast to non-const char * in host
template initialization." from Kristian Høgsberg
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Found and debugged by Jay Fenlason <fenlason@redhat.com>.
The bug was especially noticeable with direct I/O over fw-sbp2.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
This check is bogus:
- Maximum asynchronous payload size for S800...S3200 is 4096.
- The p->payload_length is totally uninteresting. Only the
request->length of the subsequently allocated and initialized
struct fw_request is of significance.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
The CPU must not touch the buffer after it was DMA-mapped.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
The CPU must not touch the buffer after it was DMA-mapped.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
- The CPU must not touch the buffer after it was DMA-mapped.
- The size argument of dma_unmap_single(...page_table...) was bogus.
- Move a comment closer to the code to which it refers to.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Add rudimentary check for the case that the page table overflows due to
merging of s/g elements by the IOMMU. This would have lead to
overwriting of arbitrary memory.
After this change I expect that an offending command will be
unsuccessfully retried until the scsi_device is taken offline by SCSI
core. It's a border case and not worth to implement a recovery
strategy.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Acked-by: Kristian Høgsberg <krh@redhat.com>
Replace a cast with a container_of(). As long as nobody reorders the
structure elements, they do the same thing, but container_of() is more
readable.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de> (added complete_command_orb)
Acked-by: Kristian Høgsberg <krh@redhat.com>
This affects of course only the "soft shutdown" case, e.g. "modprobe -r
firewire-sbp2", while it doesn't matter for hot unplug.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
This currently only affects one bridge in the hardwired blacklist.
I don't own one of those, hence haven't tested it.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
cleanup after "firewire: support S100B...S400B and link slower than PHY"
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Use a speed probe to determine the speed over 1394b buses and of nodes
which report a link speed less than their PHY speed.
Log the effective maximum speed of newly created nodes in dmesg.
Also, read the config ROM (except bus info block) at the maximum speed
rather than S100. This isn't a real optimization though because we
still only use quadlet read requests for the entire ROM.
The patch also adds support for S1600 and S3200, although such hardware
does not exist yet.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Table-based gap count optimization cannot be used if 1394b repeater PHYs
are present. But it does work with 1394b leaf nodes.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
cleanup after support of single-buffer requests was dropped
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Hoegsberg <krh@redhat.com>
Alas that won't work so good, because nobody reads help texts.
I thought about adding some crude multiple choice selection (build the
old stack, build the new stack, build both stacks). It's possible, but
it would introduce awkward dummy config variables.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
descriptor.data_address is little endian
Tested-by: Olaf Hering <olh@suse.de>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
This patch fixes an OOPS on cdev release for an fd where iso context
creation failed.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The struct is part of the userspace interface and can not use
bitfields. This patch replaces the bitfields with a __u32 'control'
word and provides access macros to set the bits.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
It's a low-impact design, that just makes suspend/resume look like
a bus reset to the upper level drivers, but it should be sufficient.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The attribute /sys/bus/scsi/devices/*:*:*:*/ieee1394_id, as generated by
the old sbp2 driver, is typically used to create persistently named
links in /dev/disk/by-id.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Fix this warning on x86-64
drivers/firewire/fw-cdev.c:798: warning: initialization from incompatible pointer type
by making the return code of ioctl_send_request() the same as all the
other ioctl_xxx() return codes.
Signed-off-by: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Of course everybody immediately associates "fw-" with FireWire, not
firmware or firewall or whatever. But "firewire-" has a nice ring to
it too.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Acked-by: Kristian Hoegsberg <krh@bitplanet.net>
need it for page_private(), not all targets have it pulled indirectly
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The SCSI layer only passes sg requests down, so drop the
use_sg == 0, request_bufflen != 0 case.
Signed-off-by: Kristian Hoegsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Avoids an extra allocation and simplifies lifetime rules for the scsi_host.
Signed-off-by: Kristian Hoegsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
This patch loads fw-sbp2 if sbp2 is still in the config file. So one can
go back and forth between releases without worry about the root
filesystem drivers.
Signed-off-by: Kristian Hoegsberg <krh@redhat.com>
Existing mkinitrd scripts still have to be adapted, unless they grok
module aliases.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Drop filenames from file preamble, drop editor annotations and
use standard indent style for block comments.
Signed-off-by: Kristian Hoegsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de> (fixed typo)
With the CRC ITU-T implementation available in lib/ we can use that instead.
This also fixes a bug in the topology map crc computation.
Signed-off-by: Kristian Hoegsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de> (fixed Kconfig)
Currently create context always returns 0 and the other iso
ioctls will expect user space to pass that in for subsequent ioctls.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Also, with this change, refactor ioctl dispatch code to do the
copying from and to user space as indicated by the IOC annotations.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Make firewire entry points not 'extern':
drivers/firewire/fw-device.c:160:25: warning: function 'fw_device_get' with external linkage has definition
drivers/firewire/fw-device.c:167:13: warning: function 'fw_device_put' with external linkage has definition
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
You don't know what type is used to implement u64.
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
(Reverted part which moved it from eh_abort_handler to eh_host_reset_handler)
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
In some situations we can receive the ORB status write before we
have received the ORB pointer write response. When this happens,
we assume that the fw_transaction is finished and free the ORB
struct containing the fw_transaction.
This fix make the status write logic only accept status writes
for ORBs where the initial ORB pointer write transaction finished.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Or the SAM status codes from the device sense data into the
command error code.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Some devices have a really long power-on cycle, and we fail to
successfully probe these if they're plugged in and then turned on.
There's really no down-side to bumping the number of retries and the
retry delay, and most devices will get picked up within the first couple
of retries anyway.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
In case the topology build fails, we want to retain the old topology
info until another reset finishes and results in a valid new tree. If
we clear card->irm_node to NULL and the topology build fails, we end up
dereferencing a NULL pointer in a few places.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Generalize the way we keep track of the various resources and
assign a unique handle to each resource.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Make event struct layout common part include the closure and add a
union for all event types; provide a mechanism for setting the
bus reset event closure.
Shuffle struct fw_cdev_queue_iso fields around to be 64-bit safe.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
We dynamically create an attribute group for the key present on the
device in hand and point device->group to it. This way the device
core adds the sysfs attributes for us as the device is added.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Marc Butler <marc@adaptivecode.com>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de> (added whitespace)
We export the entire config rom, so this is technically redundant,
but should make udev rules and HAL integration easier.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The device_type struct is useful for setting attributes for a device
and overriding .release and .uevent for a group of devices. This patch
uses it this way to clean up the sysfs code a bit.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
This serializes the update callback with the probe and remove callback from
the driver core and prevents remove from being called while update is
running for the same device.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Setting these at create_context time or start_iso time doesn't matter
much, but raw1394 sets them at start_iso time so that will be easier to
emulate this way.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The dualbuffer DMA setup did not account for the iso trailer word
and thus didn't work correctly. With this fixed we can drop the
dual buffer fallback mode.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Without this, pending transactions will dereference freed memory
if they complete after the device file has been closed.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The old async transmit context handling was starting and stopping
DMA for every packet transmission. This could cause silently failing
packet transmission, if the DMA was reprogrammed too close to being
stopped.
The general context code keeps DMA running at all times and fixes this
problem. It's also a nice cleanup.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Drop the event list semaphore and only use the wait queue and the list
to synchronize queue access. Break out of a poll or read whenever
the device is disconnected.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The mod_timer based timing out of orb was a little to agressive and
would time out legit, but long-lived scsi cmds. Besides, the scsi
stack keeps track of this already. Since we're only timing out
management orbs, go back to wait_for_completion_timeout.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The old mechanism kept a struct cdev for each fw device, but fops->release
would reference this struct after the device got freed in some cases.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Repurpose the get_config_rom ioctl to a general get_info ioctl.
This ioctl is now used for version negotiation, and optionally
returns the config rom, and the current bus info.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Use PREPARE_DELAYED_WORK to just change the function pointer.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
We can't take the klist lock for the child device list in interrupt
context.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Some flaky controllers doesn't honor the masterIntEnable bits
and can generate bus reset events even if that bit is cleared.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
This lets us break out "Juju" as the model name in the config rom.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The obsolete SA_xxx interrupt flags have been used despite the scheduled
removal. Fixup the remaining users in -mm.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
We need the channel number as we queue up iso packets for transmission
so we can fill out the header correctly.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
When a DMA descriptor is appended to the context we sync it for
DMA and the device might potentially read it immediately. So,
we can't set the IRQ bits in the descriptor after appending.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
When the DMA is setup to not strip any headers, we need to use
the buffer fill descriptor instead of the dual buffer, since the
dual buffer descriptor must strip a non-zero number of header quadlets.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The descriptor circular buffer logic used for iso transmission is
useful for async transmit too, so pull the sharable logic out in
a few standalone functions.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
This patch splits out the iso buffer so we can initialize it at mmap
time with the size provided in the mmap call. Furthermore, allocate
the backing pages using alloc_page to avoid setting up kernel side
virtual memory mappings for the pages.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
The *Clear registers returns the masked value when read which is
what we want.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
This lets the SCSI stack retry the command when a SCSI command is
interrupted by a FireWire bus reset.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
This prevents superfluous bus traffic as fw-sbp2 logs in only to
get kicked off the device by another bus reset as the driver core
does bus management. Scheduling it this way lets the driver core
finish bus management before higher level drivers get the update
callback.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Sometimes we reconnect too soon, sometimes too late. Adding a retry
mechanism make the reconnect step much more robust.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
When a management ORB times out, either because the fw_transaction
times out or when we don't get the status write, we need to properly
cancel the entire operation.
Signed-off-by: Kristian Høgsberg <krh@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Stefan Richter