usb_disk: Cancel queued transfers on the interrupt endpoint in removed().

May help with some KDLs.

usb_disk: Fix double-lock KDL on device removal.

Fixes #18431.

usb_disk: Ensure we are in a consistent locking state before calling free.

Should fix the KDL in #16745.

usb_disk: Probe by USB configuration instead of attributes.

In the case of devices where the mass-storage interface is not the
first one, we will miss it unless we were to check all attributes
one at a time in order. Instead of doing that, just fetch the configuration
and enumerate it directly.

usb_disk: Granularize locking.

There is now an "io_lock" which must be held when performing any USB
operations. All ioctls that read basic status, size, icon, etc. information
do not need to acquire it, of course, which should improve userland
lockup occurrences on congested USB disks.

usb_disk: Clean up and consolidate mutex/semaphore creation/destruction.

usb_disk: Use get_attr_uint8 instead of a loop to get class/subclass/protocol.

Suggested by axeld. The original logic was copied from the USB-ECM
driver, which may also be able to be simplified here.

usb_disk: Use DMAResource for bouncing and support physical I/O requests.

A "bypass" mechanism is left in for when DMAResource would just add
overhead for no reason. All other I/O goes through it and is submitted
to the USB stack as physical addresses.

Tested in QEMU, can still boot from USB on all busses.

Fixes #15569.

Change-Id: I26bfd2208de4ebe1a17170a7034316076927663f
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6480
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

usb_disk: Move icon data to a separate header.

usb_disk: Transition to "new" driver API.

Change-Id: Ia46cb6ddc9f83917a8f797149508d35b770e44f1
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6473
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

usb_disk: Retry actions after reset_recovery.

This is mostly important for B_GET_MEDIA_STATUS, which actually will
retry actions if the err_act is set appropriately.

Following this change, stalls are now fully recoverable without causing
the file descriptors of the mounted partition to be cleared, and the
system can resume normal operation without a problem.

Fixes part of #15569.
May help with #16745, #18185, #17543, #18421.

usb_disk: Cancel queued transfers when clearing HALTs.

XHCI, at least, has "sticky" HALT states which must be cleared at
the controller level. Invoking cancel_queued_transfers takes care
of this.

This fixes USB disks spontaneously unmounting whenever stalls occur.
However, trying to read anything from the mounted partition results
in "Bad file descriptor", probably due to media-status error propagation.

add physical_block_size field where applicable

only scsi_disk checks the actual value, other drivers take the logical block size.

This change reports the physical block size from the disk rather than the block
size used by IDE/SATA/SCSI commands. On typical modern SATA disks, the SATA
commands will use 512 byte blocks, but the disk will actually read and write
4K blocks internally. This is only of importance for partition alignment for DriveSetup,
and is independant of file systems or partitioning systems. This could also influence
the recommended block size for some file systems.

Change-Id: Id0f2e22659e89fcef64c1f8d04f81cd68995e01f
Reviewed-on: https://review.haiku-os.org/c/haiku/+/5667
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>
Reviewed-by: Adrien Destugues <pulkomandy@pulkomandy.tk>

usb_disk: add support for drives bigger than 2TB

* tested read/write with a NTFS partition at the end of a 5TB USB hard disk.
* fix #14670

Change-Id: Ic005dfb7ef94b50a2d6fc8099ef5c83ec6b4a730
Reviewed-on: https://review.haiku-os.org/c/haiku/+/5397
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

usb_disk: fix SMAP violation for B_GET_MEDIA_STATUS ioctl

should fix #17283. To be tested.

Change-Id: If594129a6e16b50c314efa2a7bb1d18e01cdad15
Reviewed-on: https://review.haiku-os.org/c/haiku/+/4505
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

usb_disk: use MutexLocker in usb_disk_ioctl()

Change-Id: Ic3994f6267ea24cf776cb067e0d94226132f4c7e
Reviewed-on: https://review.haiku-os.org/c/haiku/+/3638
Reviewed-by: Jérôme Duval <jerome.duval@gmail.com>

Minor tweaks to fix the build after the last commit.

* firewire and freebsd_network expect the macros come from
sys/param.h, as this is one of the places FreeBSD defines them
* All others are Haiku-native and can use Be-style macros.

usb_disk: Handle user buffer-to-partial buffer copies correctly.

Discovered by an SMAP violation triggered by running "writembr" on a
USB drive.

Merge usb_floppy back into usb_disk.

- USB pen drive seems to still work. More extensive testing welcome.
- USB floppies don't work yet, but they don't work anymore with the
current driver, either. I'm still investigating that part.

Fixes #9276

Change-Id: I8aa5ab828ad2ad867d0c187062d6e179372fc2ad
Reviewed-on: https://review.haiku-os.org/747
Reviewed-by: Jérôme Duval <jerome.duval@gmail.com>
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

drivers: remove some non-haiku support code.

Change-Id: Ic55bb4832adabeb807f763c87ad938e74fb3a97d
Reviewed-on: https://review.haiku-os.org/520
Reviewed-by: Axel Dörfler <axeld@pinc-software.de>
Reviewed-by: waddlesplash <waddlesplash@gmail.com>

usb_disk: add the Zip icon

not used yet, my USB Zip drive seems to have gone to the mechanical valhalla.

usb_disk: add icon for mobile phones

Add a match for my Kazam dumb phone, which is not very descriptive...

usb_disk: use SD card and Memory Stick icons when found

Check for partial matches on the product_name string.

Tested with a USB card reader and an SD card.

Feel free to add icons for Compact Flash and Secure Media and more matches.

usb_disk: use CD icon for optical media

Tested with a CD drive and a USB adapter.

usb_disk: add some icons

- CD
- SD
- MemoryStick

usb_disk, usb_floppy: missing const.

USB_massbulk.h: use a proper naming for a public header.

usb_disk: extract specs into a public header.

add-ons/kernel: add spaces between literals and identifiers

Due to introduction of user-defined suffixes C++11 requires that there
is a space between literal and identifier to avoid ambiguity.

This patch makes the whole kernel build successfully with C++11.

usb_disk: accept B_DEV_MEDIA_CHANGED as a ready unit.

usb_disk: base the error sense handling on a reference table.

* imported asc-num.txt as a reference, was used to generate the asc sense table.
* use the sense asc and key tables to know which action and status codes are
to be applied.
* tested with an hard disk and a dvd reader.
* these tables could be reused by the scsi_periph module.

usb_disk: fixed mounting a USB disk on x86_64

* usb_disk_prepare_partial_buffer() was miscalculating the offset in the temp buffer,
which led to a page fault later when copying to the user buffer.

usb_disk: fixes 64bit trace build

usb_disk: returns B_DEV_NOT_READY for SCSI_SENSE_KEY_NOT_READY

...instead of B_DEV_NO_MEDIA. The latter seems to imply we stop testing
the unit is ready in usb_disk_device_added().
Based on a patch provided by markh in #9589.

usb_disk: fixed mounting a USB disk on x86_64

* usb_disk_prepare_partial_buffer() was miscalculating the offset in the temp buffer,
which led to a page fault later when copying to the user buffer.

usb_disk: fixes 64bit trace build

usb_disk: returns B_DEV_NOT_READY for SCSI_SENSE_KEY_NOT_READY

...instead of B_DEV_NO_MEDIA. The latter seems to imply we stop testing
the unit is ready in usb_disk_device_added().
Based on a patch provided by markh in #9589.

Include the actual error message in usb_disk debug output.

Implement B_GET_DEVICE_NAME in usb_disk returning SCSI info.

The SCSI inquiry vendor and product information is concatenated into a
single string to form the device name. Multiple spaces are then
collapsed to make the string more readable. The space padding is quite
common as the fields are fixed size in the inquiry block.

Consolidated and fixed device_geometry computation.

* The only implementation that would accept more than 2 TB was the one in
scsi_disk. But even that one was limited to 63 TB.
* Now there is a new utility function devfs_compute_geometry_size() which
does it correctly for sizes up to 2^64 which should be good enough for
quite some time :-)
* This fixes bug #8992.

Get USB functional on x86-64.

- Various changes to printf format specifiers to make them 32/64 safe.
- Adjustments to various descriptor structures since addr_t is not 32-bit
on x86-64, and consequently this breaks the physical structure that's
expected. Also adjusted padding to accomodate 32-byte boundaries when
64-bit pointers are in use.
- Adjustments to make use of phys_addr_t where needed.

Force read/write mode for "removable media" devices as well. This makes
it possible to write files to my MP3 player.

Don't try to do another request sense if the failing operation already was a
request sense. Otherwise we can easily run into an infinite recursion. Should
fix #8022.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42840 a95241bf-73f2-0310-859d-f6bbb57e9c96

Tiny cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42838 a95241bf-73f2-0310-859d-f6bbb57e9c96

small logic fix on my part, not properly handling usb drives with empty media. Potential fix to #7687

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42186 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Put uint8 back to uint32 as per ML recommendations
* Begin using multi-architecture friendly formatting macros per ML
* Change uint32 snoozeTime to preferred bigtime_t per ML
* Thanks for the great feedback!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42062 a95241bf-73f2-0310-859d-f6bbb57e9c96

change to non-posix int introduced a small bug, sorry about that

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42060 a95241bf-73f2-0310-859d-f6bbb57e9c96

small change, use non-posix style int typedef

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42059 a95241bf-73f2-0310-859d-f6bbb57e9c96

* change overly large uint32 to uint8 as we only count to 12 max
* add additional tracing to usb_disk lun inquiry
* cd and other optical disk devices can take some time to spin up
when plugged. Now we give them some time to do so to ensure we
can detect the media type properly if there is a cd in the drive.
Things non-mechanical (flash drives) will continue almost immediately
as there is no spin up time. The more attempts, the longer the wait.
We now try 7 times (28 seconds total), my modern usb cdrom takes
about 6 (21 seconds) of those cycles to be ready on power up and 4
(10 seconds) on media change while powered up.
* tested with usb media and cd media. Doesn't introduce any visible
delay to the user.
* the write protected status in lun is now correct for my usb cd device
on plug (it wasn't previously)
* still seeing it though as read/write. Looking at this now.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42058 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix dereference after free. CID 10543.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@39909 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix some coding style violations (case scope blocks get their
opening parenthesis on a new line & 80 chars/line limit).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@38837 a95241bf-73f2-0310-859d-f6bbb57e9c96

Patch by Matthias: The residue field is not maintained
correctly by all USB devices, so calculate it instead.
Gets some USB disk devices working like the SuperTOP.
MacOS X is doing the same thing, as researched by Jérôme.
Closes ticket #6604. Thanks!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@38836 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implement support for ATAPI command set devices. It's used by some USB CD drives
and easy to implement since ATAPI basically is SCSI anyway and we don't use many
problematic commands. Only tested for the 0x05 subclass, but 0x02 should work
the same.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36162 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Add a missing result assign that would cause the synchronize ioctl to always
return an error.
* Properly use the name length instead of a hardcoded buffer size when composing
the name of the raw device and ensure proper termination.
* Case new ioctls for Haiku as the target platform. Indeed this driver works
fine on BeOS even though it was written natively for Haiku.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33240 a95241bf-73f2-0310-859d-f6bbb57e9c96

Also test for the "no medium persent" case and don't tell B_DEV_MEDIA_CHANGED in
such a case as that would try to read info about a non-present medium.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32862 a95241bf-73f2-0310-859d-f6bbb57e9c96

Actually don't report any failures of the TUR command, as for devices with no
media present they will always fail with the no media or media changed errors.
+alphabranch


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32732 a95241bf-73f2-0310-859d-f6bbb57e9c96

If the device tells us that it doesn't support the Test Unit Ready (TUR) command,
don't try to use it for media status polling. In those cases we'll assume a
fixed device with no exchangable medias and therefore always return B_OK.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32731 a95241bf-73f2-0310-859d-f6bbb57e9c96

Enlarge the timeout for data transfers. It was 2 seconds before and under a bit
of load, when using large enough block sizes or when simply having a slow device
this is by far not enough. It is now at 15 seconds, which should reduce timeout
problems to those cases where the device actually get's stuck (because of us
doing something wrong).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32615 a95241bf-73f2-0310-859d-f6bbb57e9c96

Do not base the device number on the device count. Fixes bug #4267.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32474 a95241bf-73f2-0310-859d-f6bbb57e9c96

Adding mode sense command to retrieve the write protected status. Doesn't work
yet, so not enabled for now.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30082 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix missing mutex_unlock() in case of opening an already removed device.
Spotted by Mika Lindqvist, thanks!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28957 a95241bf-73f2-0310-859d-f6bbb57e9c96

Style cleanup. Please at least try to...


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28953 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implement B_GET_ICON_NAME to return "devices/drive-removable-media-usb" as it seems correct as per the FreeDesktop.org Icon Naming Specification we agreed to use for now, cf. http://standards.freedesktop.org/icon-naming-spec/icon-naming-spec-latest.html


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28949 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add ioctl cases for B_GET_ICON, B_GET_ICON_NAME and B_GET_VECTOR_ICON. The first
won't be implemented, the second currently isn't and the third is. This gets
rid of the frequent "unhandled ioctl" messages when using USB drives and also
adds the nice pendrive icon to the mount list when using USB storage devices.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28948 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fix a lock leak in the eject/load ioctls. It did return directly without
unlocking the device again, causing deadlocks after unmounting a USB mass
storage device.
* Synchronize on close again, but this time with proper locking of the device.
* Restructured usb_disk_synchronize() a bit and updated comments.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28934 a95241bf-73f2-0310-859d-f6bbb57e9c96

* The close hook did trigger a synchronize without locking the device. This
could have messed up the state of other transfers currently running on that
device. Since devices are regularly opened/closed for enumeration/scanning
from different threads, this could've easily lead to bad situations. I've
removed the sync completely as it's not our task to issue it and because
a close doesn't always correspond with an unmount at all.
* Retry receiving the command status wrapper also when another error than a
stall is returned. The specs aren't too specific, but the graphic suggests
this is a general recovery path.
* Do a reset in case there is an error during data transfer to start the next
command from a clean state.
* Make sure we never acknowledge more data than we actually transfered. This
is to make sure devices that return broken residue values do not mess up our
transfers.
* Detect a few more cases of invalid and non-meaningful command status wrappers.
* If the device explicitly tells us that the sync command isn't supported don't
try it a few more times. Only retry at most five times if an unspecific error
is returned that could also come from another (temporary) error case.
* Add a bit more trace output.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28930 a95241bf-73f2-0310-859d-f6bbb57e9c96

While the comment was correct with regards to collecting the semaphore after
the cancel, it wasn't actually done. This could bring a device out of sync in
the case timeouts actually happened (which shouldn't be a commen case).
Fixed that and increased the timeout to 2 seconds in favor of slower devices.
Might need some fine tuning later still.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26197 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add comment explaining the reason for retrying capacity reads.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26083 a95241bf-73f2-0310-859d-f6bbb57e9c96

Impose a one second timeout on all usb_disk data transfers. This should avoid
hanging systems on boot, but probably just hides a problem somewhere else, as
the transfers should timeout on their own if the device doesn't respond.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26082 a95241bf-73f2-0310-859d-f6bbb57e9c96

Retry reading the capacity up to three times. The first try might only yield a
unit attention telling us that the device or media status changed, which is
expected.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26081 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implement B_EJECT_DEVICE and B_LOAD_MEDIA through the START_STOP_UNIT SCSI
command.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25955 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix a deadlock when removing a mass storage device that had active transfers.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25829 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Replaced all instances of benaphores in the kernel code by mutexes.
* Removed kernel benaphores.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25690 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Updating device->removed must be protected by the device lock
* Also updating device->open_count must be protected by the device list lock

This fixes the (unlikely) case where a device was removed from usb_disk
standpoint while it still had an open read/write/ioctl operation.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24811 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implement a downcounter for the synchronize cache command. There are devices
that do not support it (probably because they do not have a cache at all like
in flash media) and this should at least avoid spamming the syslog when this is
the case. It will try 5 times and then disable syncing for that device. Untested
as of yet though.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24803 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Add support for devices with multiple LUNs
* A device now has one or more LUNs which are used for LUN specific functions
* Adapted these LUN specific functions to the new design and also change the
(un-)publishing accordingly
* Handle "no media" cases more correctly
* Output some interesting device infos also with tracing disabled
* Add some more helpful debug output here and there

Devices that provide multiple logical units (like those n in 1 card readers)
should now be usable with usb_disk too. Also we don't fail when a device is
attached which currently does not contain a media. Instead handle media changes
correctly so inserting a media works as expected. Note that hot-unplugging and
media removal still does not quite work right at the devfs or disk device
manager level and easily crashes the system.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24799 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Retry some times when the device reports it is not ready
* Use the CBW_SIGNATURE define instead of the numeric value

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24729 a95241bf-73f2-0310-859d-f6bbb57e9c96

Introduce a simple usb_disk driver that supports USB mass storage devices of
the bulk-only class using transparent SCSI commands (i.e. most of the current
external harddrives and flash media). It emulates the few SCSI commands needed
to get this sort of devices working and does not interface with the SCSI
subsystem. This makes it far easier to get working and also far better fits
how the USB stack works (as drivers can be dynamically rescanned when device
changes occur). This will allow for easy dynamic un- and replugging at runtime.
Note that while the device is currently published when you plug something in at
runtime, the partitions are not scanned and published automatically. It works
however if you have the device plugged in at boot.
I haven't added this driver to the image yet, but if properly installed and
with boot module links to the driver and USB stack, it allows to boot Haiku
from USB media like memory sticks or external harddrives.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24713 a95241bf-73f2-0310-859d-f6bbb57e9c96

usb_disk, usb_floppy: missing const.

USB_massbulk.h: use a proper naming for a public header.

usb_disk: extract specs into a public header.

add-ons/kernel: add spaces between literals and identifiers

Due to introduction of user-defined suffixes C++11 requires that there
is a space between literal and identifier to avoid ambiguity.

This patch makes the whole kernel build successfully with C++11.

usb_disk: accept B_DEV_MEDIA_CHANGED as a ready unit.

usb_disk: base the error sense handling on a reference table.

* imported asc-num.txt as a reference, was used to generate the asc sense table.
* use the sense asc and key tables to know which action and status codes are
to be applied.
* tested with an hard disk and a dvd reader.
* these tables could be reused by the scsi_periph module.

usb_disk: fixed mounting a USB disk on x86_64

* usb_disk_prepare_partial_buffer() was miscalculating the offset in the temp buffer,
which led to a page fault later when copying to the user buffer.

usb_disk: fixes 64bit trace build

usb_disk: returns B_DEV_NOT_READY for SCSI_SENSE_KEY_NOT_READY

...instead of B_DEV_NO_MEDIA. The latter seems to imply we stop testing
the unit is ready in usb_disk_device_added().
Based on a patch provided by markh in #9589.

usb_disk: fixed mounting a USB disk on x86_64

* usb_disk_prepare_partial_buffer() was miscalculating the offset in the temp buffer,
which led to a page fault later when copying to the user buffer.

usb_disk: fixes 64bit trace build

usb_disk: returns B_DEV_NOT_READY for SCSI_SENSE_KEY_NOT_READY

...instead of B_DEV_NO_MEDIA. The latter seems to imply we stop testing
the unit is ready in usb_disk_device_added().
Based on a patch provided by markh in #9589.

Include the actual error message in usb_disk debug output.

Implement B_GET_DEVICE_NAME in usb_disk returning SCSI info.

The SCSI inquiry vendor and product information is concatenated into a
single string to form the device name. Multiple spaces are then
collapsed to make the string more readable. The space padding is quite
common as the fields are fixed size in the inquiry block.

Consolidated and fixed device_geometry computation.

* The only implementation that would accept more than 2 TB was the one in
scsi_disk. But even that one was limited to 63 TB.
* Now there is a new utility function devfs_compute_geometry_size() which
does it correctly for sizes up to 2^64 which should be good enough for
quite some time :-)
* This fixes bug #8992.

Get USB functional on x86-64.

- Various changes to printf format specifiers to make them 32/64 safe.
- Adjustments to various descriptor structures since addr_t is not 32-bit
on x86-64, and consequently this breaks the physical structure that's
expected. Also adjusted padding to accomodate 32-byte boundaries when
64-bit pointers are in use.
- Adjustments to make use of phys_addr_t where needed.

Force read/write mode for "removable media" devices as well. This makes
it possible to write files to my MP3 player.

Don't try to do another request sense if the failing operation already was a
request sense. Otherwise we can easily run into an infinite recursion. Should
fix #8022.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42840 a95241bf-73f2-0310-859d-f6bbb57e9c96

Tiny cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42838 a95241bf-73f2-0310-859d-f6bbb57e9c96

small logic fix on my part, not properly handling usb drives with empty media. Potential fix to #7687

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42186 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Put uint8 back to uint32 as per ML recommendations
* Begin using multi-architecture friendly formatting macros per ML
* Change uint32 snoozeTime to preferred bigtime_t per ML
* Thanks for the great feedback!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42062 a95241bf-73f2-0310-859d-f6bbb57e9c96

change to non-posix int introduced a small bug, sorry about that

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42060 a95241bf-73f2-0310-859d-f6bbb57e9c96

small change, use non-posix style int typedef

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42059 a95241bf-73f2-0310-859d-f6bbb57e9c96

* change overly large uint32 to uint8 as we only count to 12 max
* add additional tracing to usb_disk lun inquiry
* cd and other optical disk devices can take some time to spin up
when plugged. Now we give them some time to do so to ensure we
can detect the media type properly if there is a cd in the drive.
Things non-mechanical (flash drives) will continue almost immediately
as there is no spin up time. The more attempts, the longer the wait.
We now try 7 times (28 seconds total), my modern usb cdrom takes
about 6 (21 seconds) of those cycles to be ready on power up and 4
(10 seconds) on media change while powered up.
* tested with usb media and cd media. Doesn't introduce any visible
delay to the user.
* the write protected status in lun is now correct for my usb cd device
on plug (it wasn't previously)
* still seeing it though as read/write. Looking at this now.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42058 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix dereference after free. CID 10543.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@39909 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix some coding style violations (case scope blocks get their
opening parenthesis on a new line & 80 chars/line limit).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@38837 a95241bf-73f2-0310-859d-f6bbb57e9c96

Patch by Matthias: The residue field is not maintained
correctly by all USB devices, so calculate it instead.
Gets some USB disk devices working like the SuperTOP.
MacOS X is doing the same thing, as researched by Jérôme.
Closes ticket #6604. Thanks!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@38836 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implement support for ATAPI command set devices. It's used by some USB CD drives
and easy to implement since ATAPI basically is SCSI anyway and we don't use many
problematic commands. Only tested for the 0x05 subclass, but 0x02 should work
the same.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36162 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Add a missing result assign that would cause the synchronize ioctl to always
return an error.
* Properly use the name length instead of a hardcoded buffer size when composing
the name of the raw device and ensure proper termination.
* Case new ioctls for Haiku as the target platform. Indeed this driver works
fine on BeOS even though it was written natively for Haiku.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33240 a95241bf-73f2-0310-859d-f6bbb57e9c96

Also test for the "no medium persent" case and don't tell B_DEV_MEDIA_CHANGED in
such a case as that would try to read info about a non-present medium.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32862 a95241bf-73f2-0310-859d-f6bbb57e9c96

Actually don't report any failures of the TUR command, as for devices with no
media present they will always fail with the no media or media changed errors.
+alphabranch


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32732 a95241bf-73f2-0310-859d-f6bbb57e9c96

If the device tells us that it doesn't support the Test Unit Ready (TUR) command,
don't try to use it for media status polling. In those cases we'll assume a
fixed device with no exchangable medias and therefore always return B_OK.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32731 a95241bf-73f2-0310-859d-f6bbb57e9c96

Enlarge the timeout for data transfers. It was 2 seconds before and under a bit
of load, when using large enough block sizes or when simply having a slow device
this is by far not enough. It is now at 15 seconds, which should reduce timeout
problems to those cases where the device actually get's stuck (because of us
doing something wrong).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32615 a95241bf-73f2-0310-859d-f6bbb57e9c96

Do not base the device number on the device count. Fixes bug #4267.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32474 a95241bf-73f2-0310-859d-f6bbb57e9c96

Adding mode sense command to retrieve the write protected status. Doesn't work
yet, so not enabled for now.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30082 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix missing mutex_unlock() in case of opening an already removed device.
Spotted by Mika Lindqvist, thanks!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28957 a95241bf-73f2-0310-859d-f6bbb57e9c96

Style cleanup. Please at least try to...


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28953 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implement B_GET_ICON_NAME to return "devices/drive-removable-media-usb" as it seems correct as per the FreeDesktop.org Icon Naming Specification we agreed to use for now, cf. http://standards.freedesktop.org/icon-naming-spec/icon-naming-spec-latest.html


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28949 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add ioctl cases for B_GET_ICON, B_GET_ICON_NAME and B_GET_VECTOR_ICON. The first
won't be implemented, the second currently isn't and the third is. This gets
rid of the frequent "unhandled ioctl" messages when using USB drives and also
adds the nice pendrive icon to the mount list when using USB storage devices.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28948 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fix a lock leak in the eject/load ioctls. It did return directly without
unlocking the device again, causing deadlocks after unmounting a USB mass
storage device.
* Synchronize on close again, but this time with proper locking of the device.
* Restructured usb_disk_synchronize() a bit and updated comments.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28934 a95241bf-73f2-0310-859d-f6bbb57e9c96

* The close hook did trigger a synchronize without locking the device. This
could have messed up the state of other transfers currently running on that
device. Since devices are regularly opened/closed for enumeration/scanning
from different threads, this could've easily lead to bad situations. I've
removed the sync completely as it's not our task to issue it and because
a close doesn't always correspond with an unmount at all.
* Retry receiving the command status wrapper also when another error than a
stall is returned. The specs aren't too specific, but the graphic suggests
this is a general recovery path.
* Do a reset in case there is an error during data transfer to start the next
command from a clean state.
* Make sure we never acknowledge more data than we actually transfered. This
is to make sure devices that return broken residue values do not mess up our
transfers.
* Detect a few more cases of invalid and non-meaningful command status wrappers.
* If the device explicitly tells us that the sync command isn't supported don't
try it a few more times. Only retry at most five times if an unspecific error
is returned that could also come from another (temporary) error case.
* Add a bit more trace output.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28930 a95241bf-73f2-0310-859d-f6bbb57e9c96

While the comment was correct with regards to collecting the semaphore after
the cancel, it wasn't actually done. This could bring a device out of sync in
the case timeouts actually happened (which shouldn't be a commen case).
Fixed that and increased the timeout to 2 seconds in favor of slower devices.
Might need some fine tuning later still.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26197 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add comment explaining the reason for retrying capacity reads.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26083 a95241bf-73f2-0310-859d-f6bbb57e9c96

Impose a one second timeout on all usb_disk data transfers. This should avoid
hanging systems on boot, but probably just hides a problem somewhere else, as
the transfers should timeout on their own if the device doesn't respond.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26082 a95241bf-73f2-0310-859d-f6bbb57e9c96

Retry reading the capacity up to three times. The first try might only yield a
unit attention telling us that the device or media status changed, which is
expected.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26081 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implement B_EJECT_DEVICE and B_LOAD_MEDIA through the START_STOP_UNIT SCSI
command.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25955 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix a deadlock when removing a mass storage device that had active transfers.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25829 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Replaced all instances of benaphores in the kernel code by mutexes.
* Removed kernel benaphores.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25690 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Updating device->removed must be protected by the device lock
* Also updating device->open_count must be protected by the device list lock

This fixes the (unlikely) case where a device was removed from usb_disk
standpoint while it still had an open read/write/ioctl operation.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24811 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implement a downcounter for the synchronize cache command. There are devices
that do not support it (probably because they do not have a cache at all like
in flash media) and this should at least avoid spamming the syslog when this is
the case. It will try 5 times and then disable syncing for that device. Untested
as of yet though.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24803 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Add support for devices with multiple LUNs
* A device now has one or more LUNs which are used for LUN specific functions
* Adapted these LUN specific functions to the new design and also change the
(un-)publishing accordingly
* Handle "no media" cases more correctly
* Output some interesting device infos also with tracing disabled
* Add some more helpful debug output here and there

Devices that provide multiple logical units (like those n in 1 card readers)
should now be usable with usb_disk too. Also we don't fail when a device is
attached which currently does not contain a media. Instead handle media changes
correctly so inserting a media works as expected. Note that hot-unplugging and
media removal still does not quite work right at the devfs or disk device
manager level and easily crashes the system.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24799 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Retry some times when the device reports it is not ready
* Use the CBW_SIGNATURE define instead of the numeric value

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24729 a95241bf-73f2-0310-859d-f6bbb57e9c96

Introduce a simple usb_disk driver that supports USB mass storage devices of
the bulk-only class using transparent SCSI commands (i.e. most of the current
external harddrives and flash media). It emulates the few SCSI commands needed
to get this sort of devices working and does not interface with the SCSI
subsystem. This makes it far easier to get working and also far better fits
how the USB stack works (as drivers can be dynamically rescanned when device
changes occur). This will allow for easy dynamic un- and replugging at runtime.
Note that while the device is currently published when you plug something in at
runtime, the partitions are not scanned and published automatically. It works
however if you have the device plugged in at boot.
I haven't added this driver to the image yet, but if properly installed and
with boot module links to the driver and USB stack, it allows to boot Haiku
from USB media like memory sticks or external harddrives.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24713 a95241bf-73f2-0310-859d-f6bbb57e9c96