- 19 Apr, 2010 3 commits
-
-
Jeff Mahoney authored
Since tmpfs has no persistent storage, it pins all its dentries in memory so they have d_count=1 when other file systems would have d_count=0. ->lookup is only used to create new dentries. If the caller doesn't instantiate it, it's freed immediately at dput(). ->readdir reads directly from the dcache and depends on the dentries being hashed. When an ecryptfs mount is mounted, it associates the lower file and dentry with the ecryptfs files as they're accessed. When it's umounted and destroys all the in-memory ecryptfs inodes, it fput's the lower_files and d_drop's the lower_dentries. Commit 4981e081 added this and a d_delete in 2008 and several months later commit caeeeecf removed the d_delete. I believe the d_drop() needs to be removed as well. The d_drop effectively hides any file that has been accessed via ecryptfs from the underlying tmpfs since it depends on it being hashed for it to be accessible. I've removed the d_drop on my development node and see no ill effects with basic testing on both tmpfs and persistent storage. As a side effect, after ecryptfs d_drops the dentries on tmpfs, tmpfs BUGs on umount. This is due to the dentries being unhashed. tmpfs->kill_sb is kill_litter_super which calls d_genocide to drop the reference pinning the dentry. It skips unhashed and negative dentries, but shrink_dcache_for_umount_subtree doesn't. Since those dentries still have an elevated d_count, we get a BUG(). This patch removes the d_drop call and fixes both issues. This issue was reported at: https://bugzilla.novell.com/show_bug.cgi?id=567887Reported-by: Árpád Bíró <biroa@demasz.hu> Signed-off-by: Jeff Mahoney <jeffm@suse.com> Cc: Dustin Kirkland <kirkland@canonical.com> Cc: stable@kernel.org Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
-
Christian Pulvermacher authored
If the lower file system driver has extended attributes disabled, ecryptfs' own access functions return -ENOSYS instead of -EOPNOTSUPP. This breaks execution of programs in the ecryptfs mount, since the kernel expects the latter error when checking for security capabilities in xattrs. Signed-off-by: Christian Pulvermacher <pulvermacher@gmx.de> Cc: stable@kernel.org Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
-
Tyler Hicks authored
Create a getattr handler for eCryptfs symlinks that is capable of reading the lower target and decrypting its path. Prior to this patch, a stat's st_size field would represent the strlen of the encrypted path, while readlink() would return the strlen of the decrypted path. This could lead to confusion in some userspace applications, since the two values should be equal. https://bugs.launchpad.net/bugs/524919Reported-by: Loïc Minier <loic.minier@canonical.com> Cc: stable@kernel.org Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
-
- 23 Mar, 2010 4 commits
-
-
Tyler Hicks authored
The ecryptfs_encrypted_view mount option provides a unified way of viewing encrypted eCryptfs files. If the metadata is stored in a xattr, the metadata is moved to the file header when the file is read inside the eCryptfs mount. Because of this, we should strip the ECRYPTFS_METADATA_IN_XATTR flag from the header's flag section. This allows eCryptfs to treat the file as an eCryptfs file with a header at the front. Reviewed-by: Eric Sandeen <sandeen@redhat.com> Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
-
Tyler Hicks authored
We initially read in the first PAGE_CACHE_SIZE of a file to if the eCryptfs header marker can be found. If it isn't found and ecryptfs_xattr_metadata was given as a mount option, then the user.ecryptfs xattr is read into the same buffer. Since the data from the first page of the file wasn't cleared, it is possible that we think we've found a second tag 3 or tag 1 packet and then error out after the packet contents aren't as expected. This patch clears the buffer before filling it with metadata from the user.ecryptfs xattr. Reviewed-by: Eric Sandeen <sandeen@redhat.com> Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
-
Tyler Hicks authored
This patch renames the num_header_bytes_at_front variable to metadata_size since it now contains the max size of the metadata. Reviewed-by: Eric Sandeen <sandeen@redhat.com> Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
-
Tyler Hicks authored
Fixes regression in 8faece5f When using the ecryptfs_xattr_metadata mount option, eCryptfs stores the metadata (normally stored at the front of the file) in the user.ecryptfs xattr. This causes ecryptfs_crypt_stat.num_header_bytes_at_front to be 0, since there is no header data at the front of the file. This results in too much memory being requested and ENOMEM being returned from ecryptfs_write_metadata(). This patch fixes the problem by using the num_header_bytes_at_front variable for specifying the max size of the metadata, despite whether it is stored in the header or xattr. Reviewed-by: Eric Sandeen <sandeen@redhat.com> Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
-
- 20 Mar, 2010 3 commits
-
-
Linus Torvalds authored
-
git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6Linus Torvalds authored
* git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6: serial: sh-sci: remove duplicated #include sh: Export uncached helper symbols. sh: Fix up NUMA build for 29-bit. serial: sh-sci: Fix build failure for non-sh architectures. sh: Fix up uncached offset for legacy 29-bit mode. sh: Support CPU affinity masks for INTC controllers.
-
git://git.kernel.org/pub/scm/linux/kernel/git/dvrabel/uwbLinus Torvalds authored
* 'for-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/dvrabel/uwb: uwb: remove duplicate cpu_to_le16() uwb: declare MODULE_FIRMWARE() in i1480 DFU driver uwb: make USB device id table constant uwb: wlp: refactor wlp_get_<attribute>() macros
-
- 19 Mar, 2010 30 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/mattst88/alpha-2.6Linus Torvalds authored
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mattst88/alpha-2.6: alpha: fix compile errors in dma-mapping-common.h alpha: remove trailing spaces in messages alpha: use __ratelimit
-
git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpcLinus Torvalds authored
* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: powerpc: Remove IOMMU_VMERGE config option powerpc: Fix swiotlb to respect the boot option powerpc: Do not call prink when CONFIG_PRINTK is not defined powerpc: Use correct ccr bit for syscall error status powerpc/fsl-booke: Get coherent bit from PTE powerpc/85xx: Make sure lwarx hint isn't set on ppc32
-
git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/misc-2.6Linus Torvalds authored
* 'zerolen' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/misc-2.6: Delete zero-length file drivers/mtd/maps/omap_nor.c
-
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty-2.6Linus Torvalds authored
* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty-2.6: tty_port,usb-console: Fix usb serial console open/close regression tty: cpm_uart: use resource_size() tty_buffer: Fix distinct type warning hvc_console: Fix race between hvc_close and hvc_remove uartlite: Fix build on sparc. tty: Take a 256 byte padding into account when buffering below sub-page units Revert "tty: Add a new VT mode which is like VT_PROCESS but doesn't require a VT_RELDISP ioctl call"
-
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core-2.6Linus Torvalds authored
* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core-2.6: driver core: numa: fix BUILD_BUG_ON for node_read_distance driver-core: document ERR_PTR() return values kobject: documentation: Update to refer to kset-example.c. sysdev: the cpu probe/release attributes should be sysdev_class_attributes kobject: documentation: Fix erroneous example in kobject doc. driver-core: fix missing kernel-doc in firmware_class Driver core: Early platform kernel-doc update sysfs: fix sysfs lockdep warning in mlx4 code sysfs: fix sysfs lockdep warning in infiniband code sysfs: fix sysfs lockdep warning in ipmi code sysfs: Initialised pci bus legacy_mem field before use sysfs: use sysfs_bin_attr_init in firmware class driver
-
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6Linus Torvalds authored
* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6: (45 commits) USB: gadget/multi: cdc_do_config: remove redundant check usb: r8a66597-hcd: fix removed from an attached hub USB: xhci: Make endpoint interval debugging clearer. USB: Fix usb_fill_int_urb for SuperSpeed devices USB: cp210x: Remove double usb_control_msg from cp210x_set_config USB: Remove last bit of CONFIG_USB_BERRY_CHARGE USB: gadget: add gadget controller number for s3c-hsotg driver USB: ftdi_sio: Fix locking for change_speed() function USB: g_mass_storage: fixed module name in Kconfig USB: gadget: f_mass_storage::fsg_bind(): fix error handling USB: g_mass_storage: fix section mismatch warnings USB: gadget: fix Blackfin builds after gadget cleansing USB: goku_udc: remove potential null dereference USB: option.c: Add Pirelli VID/PID and indicate Pirelli's modem interface is 0xff USB: serial: Fix module name typo for qcaux Kconfig entry. usb: cdc-wdm: Fix deadlock between write and resume usb: cdc-wdm: Fix order in disconnect and fix locking usb: cdc-wdm:Fix loss of data due to autosuspend usb: cdc-wdm: Fix submission of URB after suspension usb: cdc-wdm: Fix race between disconnect and debug messages ...
-
Jeff Garzik authored
The content was deleted in cc87edb1, but the file remained as a zero-length file. Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
-
git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-clientLinus Torvalds authored
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client: (205 commits) ceph: update for write_inode API change ceph: reset osd after relevant messages timed out ceph: fix flush_dirty_caps race with caps migration ceph: include migrating caps in issued set ceph: fix osdmap decoding when pools include (removed) snaps ceph: return EBADF if waiting for caps on closed file ceph: set osd request message front length correctly ceph: reset front len on return to msgpool; BUG on mismatched front iov ceph: fix snaptrace decoding on cap migration between mds ceph: use single osd op reply msg ceph: reset bits on connection close ceph: remove bogus mds forward warning ceph: remove fragile __map_osds optimization ceph: fix connection fault STANDBY check ceph: invalidate_authorizer without con->mutex held ceph: don't clobber write return value when using O_SYNC ceph: fix client_request_forward decoding ceph: drop messages on unregistered mds sessions; cleanup ceph: fix comments, locking in destroy_inode ceph: move dereference after NULL test ... Fix trivial conflicts in Documentation/ioctl/ioctl-number.txt
-
git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6Linus Torvalds authored
* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6: cifs: trivial white space [CIFS] checkpatch cleanup cifs: add cifs_revalidate_file cifs: add a CIFSSMBUnixQFileInfo function cifs: add a CIFSSMBQFileInfo function cifs: overhaul cifs_revalidate and rename to cifs_revalidate_dentry
-
Peter Korsgaard authored
cdc_do_config() had a double ret check after fsg_add(). Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Yoshihiro Shimoda authored
fix the problem that when a USB hub is attached to the r8a66597-hcd and a device is removed from that hub, it's likely that a kernel panic follows. Reported-by: Markus Pietrek <Markus.Pietrek@emtrion.de> Signed-off-by: Yoshihiro Shimoda <shimoda.yoshihiro@renesas.com> Cc: stable <stable@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Sarah Sharp authored
The xHCI hardware can only handle polling intervals that are a power of two. When we add a new endpoint during a bandwidth allocation, and the polling interval is rounded down to a power of two, print the original polling interval in the endpoint descriptor. Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Matthew Wilcox authored
USB 3 and Wireless USB specify a logarithmic encoding of the endpoint interval that matches the USB 2 specification. usb_fill_int_urb() didn't know that and was filling in the interval as if it was USB 1.1. Fix usb_fill_int_urb() for SuperSpeed devices, but leave the wireless case alone, because David Vrabel wants to keep the old encoding. Update the struct urb kernel doc to note that SuperSpeed URBs must have urb->interval specified in microframes. Add a missing break statement in the usb_submit_urb() interrupt URB checking, since wireless USB and SuperSpeed USB encode urb->interval differently. This allows xHCI roothubs to actually register with khubd. Signed-off-by: Matthew Wilcox <willy@linux.intel.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Michael Brunner authored
This patch removes a double usb_control_msg that sets the cp210x configuration registers a second time when calling cp210x_set_config. For data sizes >2 the second write gets corrupted. The patch has been created against 2.6.34-rc1, but all cp210x driver revisions are affected. Signed-off-by: Michael Brunner <mibru@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Christoph Egger authored
One last bit was missed while removing the USB_BERRY_CHARGE config option in a8d4211f which gets dropped by this patch. Signed-off-by: Christoph Egger <siccegge@stud.informatik.uni-erlangen.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Maurus Cuelenaere authored
This prevents some drivers from complaining that no bcdDevice id was set. Signed-off-by: Maurus Cuelenaere <mcuelenaere@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Alessio Igor Bogani authored
The change_speed() function should be serialized against multiple calls. Use the cfg_lock mutex to do this. Signed-off-by: Alessio Igor Bogani <abogani@texware.it> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Michal Nazarewicz authored
The Kconfig help message for Mass Storage Gadget claimed the module will be named "g_file_storage" whereas it should be "g_mass_storage". Signed-off-by: Michal Nazarewicz <m.nazarewicz@samsung.com> Cc: Kyungmin Park <kyungmin.park@samsung.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Peter Korsgaard authored
Contrary to the comment in fsg_add, fsg_bind calls fsg_unbind on errors, which decreases refcount and frees the fsg_dev structure, causing trouble when fsg_add does the same. Fix it by simply leaving up cleanup to fsg_add(). Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk> Acked-by: Michal Nazarewicz <m.nazarewicz@samsung.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Mike Frysinger authored
The recent commit (0e530b45) that moved usb_ep_autoconfig from the __devinit section to the __init section missed the mass storage device. Its fsg_bind() function uses the usb_ep_autoconfig() function from non __init context leading to: WARNING: drivers/usb/gadget/g_mass_storage.o(.text): Section mismatch in reference from the function _fsg_bind() to the function .init.text:_usb_ep_autoconfig() So move fsg_bind() into __init as well. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Mike Frysinger authored
The recent change to clean out dead gadget drivers (90f79768) missed the call to gadget_is_musbhsfc() behind CONFIG_BLACKFIN. This causes Blackfin gadget builds to fail since the function no longer exists anywhere. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Dan Carpenter authored
"dev" is always null here. In the end it's only used to get the pci_name() of "pdev" which is redundant information and so I removed it. Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
spark authored
Signed-off-by: spark <spark@bandrich.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Stefan Schmidt authored
The module is called qcaux and not moto_modem. Also use help instead of ---help-- to be in sync with the other Kconfig entries. Signed-off-by: Stefan Schmidt <stefan@datenfreihafen.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Oliver Neukum authored
The new runtime PM scheme allows resume() to have no locks. This fixes the deadlock. Signed-off-by: Oliver Neukum <neukum@b1-systems.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Oliver Neukum authored
- as the callback can schedule work, URBs must be killed first - if the driver causes an autoresume, the caller must handle locking Signed-off-by: Oliver Neukum <neukum@b1-systems.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Oliver Neukum authored
The guarding flag must be set and tested under spinlock and cleared before the URBs are resubmitted in resume. Signed-off-by: Oliver Neukum <neukum@b1-systems.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Oliver Neukum authored
There's a window under which cdc-wdm may submit an URB to a device about to be suspended. This introduces a flag to prevent it. Signed-off-by: Oliver Neukum <neukum@b1-systems.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Oliver Neukum authored
dev_dbg() and dev_err() cannot be used to report failures that may have been caused by a device's removal Signed-off-by: Oliver Neukum <neukum@b1-systems.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-
Oliver Neukum authored
While an available response is read the device must not be autosuspended. This requires a flag dedicated to that purpose. Signed-off-by: Oliver Neukum <neukum@b1-systems.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-