summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* V4L/DVB: v4l2-mediabus.h: add two helper functionsHans Verkuil2010-06-011-0/+20
| | | | | Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: saa717x: remove obsolete s_fmt opHans Verkuil2010-06-011-11/+0
| | | | | Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: saa7127: remove obsolete g_fmt supportHans Verkuil2010-06-011-8/+0
| | | | | Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: cx18: remove old g/s_fmt from the cx18_av subdevHans Verkuil2010-06-013-34/+10
| | | | | | | cx18 has now switched over completely to the new mediabus subdev ops. Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: cx18: add s_mbus_fmt supportHans Verkuil2010-06-011-57/+69
| | | | | Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: ivtv: convert to use s_mbus_fmtHans Verkuil2010-06-012-6/+10
| | | | | Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: saa717x: add support for s_mbus_fmtHans Verkuil2010-06-011-16/+31
| | | | | Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: cx25840: add support for s_mbus_fmtHans Verkuil2010-06-011-41/+55
| | | | | Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: saa7115: add s_mbus_fmt opHans Verkuil2010-06-011-0/+10
| | | | | Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: V4L2 Spec: Improve the VIDIOC_QUERY_DV_PRESET descriptionHans Verkuil2010-06-011-2/+4
| | | | | | | | Make explicit what should happen when the input signal is missing, unreliable or does not map to a supported preset. Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: ngene: Remove debug messageOliver Endriss2010-06-011-5/+1
| | | | | | | Remove debug message. Signed-off-by: Oliver Endriss <o.endriss@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: ngene: MSI cleanupOliver Endriss2010-06-011-1/+5
| | | | | | | MSI cleanup. Signed-off-by: Oliver Endriss <o.endriss@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: ngene: Make command timeout workaround configurableOliver Endriss2010-06-013-16/+8
| | | | | | | | Make command timeout workaround configurable, activate it for firmware version <= 17. Signed-off-by: Oliver Endriss <o.endriss@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: ngene: Implement support for MSIOliver Endriss2010-06-012-1/+34
| | | | | | | Add MSI support, may be enabled with firmware version 18. Signed-off-by: Oliver Endriss <o.endriss@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: ngene: Do not call demuxer with interrupts disabledOliver Endriss2010-06-011-14/+14
| | | | | | | It is neither a good idea nor necessary to call the demuxer with interrupts disabled. Signed-off-by: Oliver Endriss <o.endriss@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: ngene: Support new device 'Digital Devices DuoFlex S2 miniPCIe'Oliver Endriss2010-06-011-0/+15
| | | | | | | Add subsystem id DD10/DD20 'Digital Devices DuoFlex S2 miniPCIe'. Signed-off-by: Oliver Endriss <o.endriss@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: gspca - sonixj: Add webcam 0c45:60ceWarren Bosworth Focke2010-06-012-1/+2
| | | | | | Signed-off-by: Warren Bosworth Focke <wbfocke@gmail.com> Signed-off-by: Jean-François Moine <moinejf@free.fr> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: gspca - sonixj: Add information about some potential JPEG webcamsJean-François Moine2010-06-011-1/+4
| | | | | Signed-off-by: Jean-François Moine <moinejf@free.fr> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: gspca - sonixb: Have 0c45:602e handled by sonixb instead of sn9c102Jean-François Moine2010-06-012-3/+1
| | | | | | | | This change fixes the Debian Bug #579332 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=579332 Signed-off-by: Jean-François Moine <moinejf@free.fr> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: pvrusb2: Fix kernel oops on device tear-downMike Isely2010-06-011-2/+2
| | | | | | | | | | | | | | | | | pvrusb2: Delete sysfs class device as the _very_ last step, after we're sure that all driver contexts have gone away first. This is important because it appears that there isn't any protection from a struct device instance reference a deleted struct class instance. The assumption in the kernel code appears to be that the class instance is assumed to be around for the life of the device. So we can't let the class instance go away until all referencing device instances are gone; this is ensured by delaying removal of the class instance until after the driver contexts have all gone away. This bug has been present for a very long time but it didn't apparently become malignant until recently (probably because of other changes in the kernel). Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: pvrusb2: Fix minor internal array allocationMike Isely2010-06-011-1/+1
| | | | | | | | | pvrusb2: Need one extra attribute slot allocated so that worst case still has a trailing null pointer. This wasn't causing visible symptoms; it was found through inspection while investigating other issues. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: pvrusb2: Fix USB parent device reference countMike Isely2010-06-011-2/+9
| | | | | | | | | pvrusb2: Correctly reference count pointer to parent USB device when linked from sysfs interface. This is technically a pretty nasty problem, however as far as I know nobody had been getting burned by it (yet). Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: pvrusb2: Fix kernel oops at device unregistrationMike Isely2010-06-011-3/+12
| | | | | | | | | | | | pvrusb2: Fix oops caused by touching deleted memory after unregistration. This bug was introduced when we had started using video_device_node_name() - that function was being called potentially after the underlying structure (referenced by that function) had been deleted. The fix rearranges things slightly so that the function is called before destruction takes place. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: pvrusb2: New feature to mark specific hardware support as experimentalMike Isely2010-06-012-0/+18
| | | | | | | | | | This adds a flag in the device attribute structure which can be used to mark support for a particular device as experimental. Any devices flagged in this way, when encountered at run-time, will generate a warning message to the kernel log. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: pvrusb2: Avoid using stack allocated buffers when performing USB I/OMike Isely2010-06-011-1/+11
| | | | | | | | Drivers shouldn't assume that the stack is DMA-safe. [mchehab@redhat.com: fix patch description] Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: pvrusb2: Fix Gotview hardware supportMike Isely2010-06-011-0/+1
| | | | | | | | pvrusb2: Fix RF tuner problem with gotview hardware - this bug was introduced when switching over to the subdev model of driver control Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: cx23885: Check register errorsAbylay Ospan2010-06-011-0/+2
| | | | | | | | | Fix kernel Oops when number of NetUP Dual DVB-S2-CI cards more than DVB_MAX_ADAPTERS limit. [mchehab@redhat.com: move the return to the proper place] Signed-off-by: Abylay Ospan <aospan@netup.ru> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: c-qcam: convert to V4L2Hans Verkuil2010-06-012-305/+331
| | | | | | | | Note: due to lack of hardware this conversion is untested. However, it is pretty straightforward so I do not expect any problems. Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* V4L/DVB: bw-qcam: convert to V4L2Hans Verkuil2010-06-013-415/+415
| | | | | | | | Note: due to lack of hardware I was not able to test this conversion. But it is pretty straightforward, so I do not expect any problems. Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* Linux 2.6.35-rc1v2.6.35-rc1Linus Torvalds2010-05-301-2/+2
| | | | .. and thus endeth the merge window.
* Merge branch 'slub/urgent' of ↵Linus Torvalds2010-05-302-29/+15
|\ | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6 * 'slub/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6: SLUB: Allow full duplication of kmalloc array for 390 slub: move kmem_cache_node into it's own cacheline
| * SLUB: Allow full duplication of kmalloc array for 390Christoph Lameter2010-05-301-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | Commit 756dee75872a2a764b478e18076360b8a4ec9045 ("SLUB: Get rid of dynamic DMA kmalloc cache allocation") makes S390 run out of kmalloc caches. Increase the number of kmalloc caches to a safe size. Cc: <stable@kernel.org> [ .33 and .34 ] Reported-by: Heiko Carstens <heiko.carstens@de.ibm.com> Tested-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
| * slub: move kmem_cache_node into it's own cachelineAlexander Duyck2010-05-242-28/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch is meant to improve the performance of SLUB by moving the local kmem_cache_node lock into it's own cacheline separate from kmem_cache. This is accomplished by simply removing the local_node when NUMA is enabled. On my system with 2 nodes I saw around a 5% performance increase w/ hackbench times dropping from 6.2 seconds to 5.9 seconds on average. I suspect the performance gain would increase as the number of nodes increases, but I do not have the data to currently back that up. Bugzilla-Reference: http://bugzilla.kernel.org/show_bug.cgi?id=15713 Cc: <stable@kernel.org> Reported-by: Alex Shi <alex.shi@intel.com> Tested-by: Alex Shi <alex.shi@intel.com> Acked-by: Yanmin Zhang <yanmin_zhang@linux.intel.com> Acked-by: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
* | Merge branch 'core-fixes-for-linus' of ↵Linus Torvalds2010-05-301-0/+7
|\ \ | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'core-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: mutex: Fix optimistic spinning vs. BKL
| * | mutex: Fix optimistic spinning vs. BKLTony Breeds2010-05-191-0/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, we can hit a nasty case with optimistic spinning on mutexes: CPU A tries to take a mutex, while holding the BKL CPU B tried to take the BLK while holding the mutex This looks like a AB-BA scenario but in practice, is allowed and happens due to the auto-release on schedule() nature of the BKL. In that case, the optimistic spinning code can get us into a situation where instead of going to sleep, A will spin waiting for B who is spinning waiting for A, and the only way out of that loop is the need_resched() test in mutex_spin_on_owner(). This patch fixes it by completely disabling spinning if we own the BKL. This adds one more detail to the extensive list of reasons why it's a bad idea for kernel code to be holding the BKL. Signed-off-by: Tony Breeds <tony@bakeyournoodle.com> Acked-by: Linus Torvalds <torvalds@linux-foundation.org> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: <stable@kernel.org> LKML-Reference: <20100519054636.GC12389@ozlabs.org> [ added an unlikely() attribute to the branch ] Signed-off-by: Ingo Molnar <mingo@elte.hu>
* | | Merge branch 'perf-fixes-for-linus' of ↵Linus Torvalds2010-05-308-16/+45
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: perf tui: Fix last use_browser problem related to .perfconfig perf symbols: Add the build id cache to the vmlinux path perf tui: Reset use_browser if stdout is not a tty ring-buffer: Move zeroing out excess in page to ring buffer code ring-buffer: Reset "real_end" when page is filled
| * \ \ Merge branch 'tip/perf/core' of ↵Ingo Molnar2010-05-292-8/+17
| |\ \ \ | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-2.6-trace into perf/urgent
| | * | | ring-buffer: Move zeroing out excess in page to ring buffer codeSteven Rostedt2010-05-252-8/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently the trace splice code zeros out the excess bytes in the page before sending it off to userspace. This is to make sure userspace is not getting anything it should not be when reading the pages, because the excess data was never initialized to zero before writing (for perfomance reasons). But the splice code has no business in doing this work, it should be done by the ring buffer. With the latest changes for recording lost events, the splice code gets it wrong anyway. Move the zeroing out of excess bytes into the ring buffer code. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
| | * | | ring-buffer: Reset "real_end" when page is filledSteven Rostedt2010-05-251-0/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The code to store the "lost events" requires knowing the real end of the page. Since the 'commit' includes the padding at the end of a page a "real_end" variable was used to keep track of the end not including the padding. If events were lost, the reader can place the count of events in the padded area if there is enough room. The bug this patch fixes is that when we fill the page we do not reset the real_end variable, and if the writer had wrapped a few times, the real_end would be incorrect. This patch simply resets the real_end if the page was filled. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
| * | | | perf tui: Fix last use_browser problem related to .perfconfigArnaldo Carvalho de Melo2010-05-271-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we moved to using ~/.perfconfig to set the value of use_browser, it changed from a boolean to an int so that the convention used for use_pager was followed. That convention is: -1: unspecified, that is what use_{browser,pager} is initialized 0: Don't use the browser (should be TUI), because was explicitely set to 0/off/false on ~/.perfconfig [tui] cmd =, or because we're redirecting the stdout to a file or piping it to some other command (!isatty()). 1: Use the TUI Some code was not properly audited and continued testing it as a boolean, this seems to be the last one. Reported-by: Frédéric Weisbecker <fweisbec@gmail.com> Tested-by: Frédéric Weisbecker <fweisbec@gmail.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Cc: Tom Zanussi <tzanussi@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
| * | | | perf symbols: Add the build id cache to the vmlinux pathArnaldo Carvalho de Melo2010-05-263-6/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | So that if the kernel DSO has a build id because record inserted it in the perf.data build id table in the header, or a BUILD_ID event was inserted in the stream, we first look at the build id cache ($HOME/.debug/). If we find it there, try to use it, allowing offline annotation in addition to 'perf report'. Reported-by: Stephane Eranian <eranian@google.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Cc: Tom Zanussi <tzanussi@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
| * | | | perf tui: Reset use_browser if stdout is not a ttyArnaldo Carvalho de Melo2010-05-262-1/+2
| |/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The newt initialization routines weren't being called because the output was a file (perf annotate > /tmp/bla) but use_browser was still 1, because ~/.perfconfig had it as 'on', so, later on newt routines segfaulted. Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Cc: Tom Zanussi <tzanussi@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
* | | | ia64: revert __node_random additionLinus Torvalds2010-05-301-17/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This partially reverts commit 4ec37de89d8c758ee8115e0e64b3f994910789ee ("[IA64] Fix build breakage"), since the commit that made it necessary got reverted earlier (see commit 35926ff5fba8, 'Revert "cpusets: randomize node rotor used in cpuset_mem_spread_node()"') Even if we ever re-introduce this, there is no reason to make __node_random be some architecture-specific function. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | | | Merge branch 'for-linus' of ↵Linus Torvalds2010-05-307-82/+500
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse: mm: export generic_pipe_buf_*() to modules fuse: support splice() reading from fuse device fuse: allow splice to move pages mm: export remove_from_page_cache() to modules mm: export lru_cache_add_*() to modules fuse: support splice() writing to fuse device fuse: get page reference for readpages fuse: use get_user_pages_fast() fuse: remove unneeded variable
| * | | | mm: export generic_pipe_buf_*() to modulesMiklos Szeredi2010-05-261-0/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is needed by fuse device code which wants to create pipe buffers. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
| * | | | fuse: support splice() reading from fuse deviceMiklos Szeredi2010-05-251-41/+187
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Allow userspace filesystem implementation to use splice() to read from the fuse device. The userspace filesystem can now transfer data coming from a WRITE request to an arbitrary file descriptor (regular file, block device or socket) without having to go through a userspace buffer. The semantics of using splice() to read messages are: 1) with a single splice() call move the whole message from the fuse device to a temporary pipe 2) read the header from the pipe and determine the message type 3a) if message is a WRITE then splice data from pipe to destination 3b) else read rest of message to userspace buffer Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
| * | | | fuse: allow splice to move pagesMiklos Szeredi2010-05-253-15/+167
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When splicing buffers to the fuse device with SPLICE_F_MOVE, try to move pages from the pipe buffer into the page cache. This allows populating the fuse filesystem's cache without ever touching the page contents, i.e. zero copy read capability. The following steps are performed when trying to move a page into the page cache: - buf->ops->confirm() to make sure the new page is uptodate - buf->ops->steal() to try to remove the new page from it's previous place - remove_from_page_cache() on the old page - add_to_page_cache_locked() on the new page If any of the above steps fail (non fatally) then the code falls back to copying the page. In particular ->steal() will fail if there are external references (other than the page cache and the pipe buffer) to the page. Also since the remove_from_page_cache() + add_to_page_cache_locked() are non-atomic it is possible that the page cache is repopulated in between the two and add_to_page_cache_locked() will fail. This could be fixed by creating a new atomic replace_page_cache_page() function. fuse_readpages_end() needed to be reworked so it works even if page->mapping is NULL for some or all pages which can happen if the add_to_page_cache_locked() failed. A number of sanity checks were added to make sure the stolen pages don't have weird flags set, etc... These could be moved into generic splice/steal code. Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
| * | | | mm: export remove_from_page_cache() to modulesMiklos Szeredi2010-05-251-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is needed to enable moving pages into the page cache in fuse with splice(..., SPLICE_F_MOVE). Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
| * | | | mm: export lru_cache_add_*() to modulesMiklos Szeredi2010-05-251-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is needed to enable moving pages into the page cache in fuse with splice(..., SPLICE_F_MOVE). Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
| * | | | fuse: support splice() writing to fuse deviceMiklos Szeredi2010-05-252-32/+148
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Allow userspace filesystem implementation to use splice() to write to the fuse device. The semantics of using splice() are: 1) buffer the message header and data in a temporary pipe 2) with a *single* splice() call move the message from the temporary pipe to the fuse device The READ reply message has the most interesting use for this, since now the data from an arbitrary file descriptor (which could be a regular file, a block device or a socket) can be tranferred into the fuse device without having to go through a userspace buffer. It will also allow zero copy moving of pages. One caveat is that the protocol on the fuse device requires the length of the whole message to be written into the header. But the length of the data transferred into the temporary pipe may not be known in advance. The current library implementation works around this by using vmplice to write the header and modifying the header after splicing the data into the pipe (error handling omitted): struct fuse_out_header out; iov.iov_base = &out; iov.iov_len = sizeof(struct fuse_out_header); vmsplice(pip[1], &iov, 1, 0); len = splice(input_fd, input_offset, pip[1], NULL, len, 0); /* retrospectively modify the header: */ out.len = len + sizeof(struct fuse_out_header); splice(pip[0], NULL, fuse_chan_fd(req->ch), NULL, out.len, flags); This works since vmsplice only saves a pointer to the data, it does not copy the data itself. Since pipes are currently limited to 16 pages and messages need to be spliced atomically, the length of the data is limited to 15 pages (or 60kB for 4k pages). Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 16:46:46 +0100