- 17 Jan, 2013 40 commits
-
-
Alex Elder authored
Add the ability to map an rbd image read-only, by specifying either "read_only" or "ro" as an option on the rbd "command line." Also allow the inverse to be explicitly specified using "read_write" or "rw". Signed-off-by:
Alex Elder <elder@inktank.com> Reviewed-by:
Yehuda Sadeh <yehuda@inktank.com> (based on commit cc0538b6)
-
Alex Elder authored
Josh proposed the following change, and I don't think I could explain it any better than he did: From: Josh Durgin <josh.durgin@inktank.com> Date: Tue, 24 Jul 2012 14:22:11 -0700 To: ceph-devel <ceph-devel@vger.kernel.org> Message-ID: <500F1203.9050605@inktank.com> From: Josh Durgin <josh.durgin@inktank.com> Right now the kernel still has one piece of rbd management duplicated from the rbd command line tool: snapshot creation. There's nothing special about snapshot creation that makes it advantageous to do from the kernel, so I'd like to remove the create_snap sysfs interface. That is, /sys/bus/rbd/devices/<id>/create_snap would be removed. Does anyone rely on the sysfs interface for creating rbd snapshots? If so, how hard would it be to replace with: rbd snap create pool/image@snap Is there any benefit to the sysfs interface that I'm missing? Josh This patch implements this proposal, removing the code that implements the "snap_create" sysfs interface for rbd images. As a result, quite a lot of other supporting code goes away. [elder@inktank.com: commented out rbd_req_sync_exec() to avoid warning] Suggested-by:Josh Durgin <josh.durgin@inktank.com> Signed-off-by:
-
Alex Elder authored
If an osd has no requests and no linger requests, __reset_osd() will just remove it with a call to __remove_osd(). That drops a reference to the osd, and therefore the osd may have been free by the time __reset_osd() returns. That function offers no indication this may have occurred, and as a result the osd will continue to be used even when it's no longer valid. Change__reset_osd() so it returns an error (ENODEV) when it deletes the osd being reset. And change __kick_osd_requests() so it returns immediately (before referencing osd again) if __reset_osd() returns *any* error. Signed-off-by:
Sage Weil <sage@inktank.com> (cherry picked from commit 685a7555) Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Sage Weil authored
Ensure that we set the err value correctly so that we do not pass a 0 value to ERR_PTR and confuse the calling code. (In particular, osd_client.c handle_map() will BUG(!newmap)). Signed-off-by:
-
Sage Weil authored
We should not set con->state to CLOSED here; that happens in ceph_fault() in the caller, where it first asserts that the state is not yet CLOSED. Avoids a BUG when the features don't match. Since the fail_protocol() has become a trivial wrapper, replace calls to it with direct calls to reset_connection(). Signed-off-by:
-
Alex Elder authored
A number of assertions in the ceph messenger are implemented with BUG_ON(), killing the system if connection's state doesn't match what's expected. At this point our state model is (evidently) not well understood enough for these assertions to trigger a BUG(). Convert all BUG_ON(con->state...) calls to be WARN_ON(con->state...) so we learn about these issues without killing the machine. We now recognize that a connection fault can occur due to a socket closure at any time, regardless of the state of the connection. So there is really nothing we can assert about the state of the connection at that point so eliminate that assertion. Reported-by:
Ugis <ugis22@gmail.com> Tested-by:
-
Alex Elder authored
When ceph_osdc_handle_map() is called to process a new osd map, kick_requests() is called to ensure all affected requests are updated if necessary to reflect changes in the osd map. This happens in two cases: whenever an incremental map update is processed; and when a full map update (or the last one if there is more than one) gets processed. In the former case, the kick_requests() call is followed immediately by a call to reset_changed_osds() to ensure any connections to osds affected by the map change are reset. But for full map updates this isn't done. Both cases should be doing this osd reset. Rather than duplicating the reset_changed_osds() call, move it into the end of kick_requests(). Signed-off-by:
-
Alex Elder authored
The kick_requests() function is called by ceph_osdc_handle_map() when an osd map change has been indicated. Its purpose is to re-queue any request whose target osd is different from what it was when it was originally sent. It is structured as two loops, one for incomplete but registered requests, and a second for handling completed linger requests. As a special case, in the first loop if a request marked to linger has not yet completed, it is moved from the request list to the linger list. This is as a quick and dirty way to have the second loop handle sending the request along with all the other linger requests. Because of the way it's done now, however, this quick and dirty solution can result in these incomplete linger requests never getting re-sent as desired. The problem lies in the fact that the second loop only arranges for a linger request to be sent if it appears its target osd has changed. This is the proper handling for *completed* linger requests (it avoids issuing the same linger request twice to the same osd). But although the linger requests added to the list in the first loop may have been sent, they have not yet completed, so they need to be re-sent regardless of whether their target osd has changed. The first required fix is we need to avoid calling __map_request() on any incomplete linger request. Otherwise the subsequent __map_request() call in the second loop will find the target osd has not changed and will therefore not re-send the request. Second, we need to be sure that a sent but incomplete linger request gets re-sent. If the target osd is the same with the new osd map as it was when the request was originally sent, this won't happen. This can be fixed through careful handling when we move these requests from the request list to the linger list, by unregistering the request *before* it is registered as a linger request. This works because a side-effect of unregistering the request is to make the request's r_osd pointer be NULL, and *that* will ensure the second loop actually re-sends the linger request. Processing of such a request is done at that point, so continue with the next one once it's been moved. Signed-off-by:
-
Alex Elder authored
In kick_requests(), we need to register the request before we unregister the linger request. Otherwise the unregister will reset the request's osd pointer to NULL. Signed-off-by:
-
Alex Elder authored
The red-black node in the ceph osd request structure is initialized in ceph_osdc_alloc_request() using rbd_init_node(). We do need to initialize this, because in __unregister_request() we call RB_EMPTY_NODE(), which expects the node it's checking to have been initialized. But rb_init_node() is apparently overkill, and may in fact be on its way out. So use RB_CLEAR_NODE() instead. For a little more background, see this commit: 4c199a93 rbtree: empty nodes have no color" Signed-off-by: -
Alex Elder authored
The red-black node node in the ceph osd event structure is not initialized in create_osdc_create_event(). Because this node can be the subject of a RB_EMPTY_NODE() call later on, we should ensure the node is initialized properly for that. Signed-off-by:
-
Alex Elder authored
The red-black node node in the ceph osd structure is not initialized in create_osd(). Because this node can be the subject of a RB_EMPTY_NODE() call later on, we should ensure the node is initialized properly for that. Add a call to RB_CLEAR_NODE() initialize it. Signed-off-by:
-
Alex Elder authored
When a connection's socket disconnects, or if there's a protocol error of some kind on the connection, a fault is signaled and the connection is reset (closed and reopened, basically). We currently get an error message on the log whenever this occurs. A ceph connection will attempt to reestablish a socket connection repeatedly if a fault occurs. This means that these error messages will get repeatedly added to the log, which is undesirable. Change the error message to be a warning, so they don't get logged by default. Signed-off-by:
-
Alex Elder authored
A connection's socket can close for any reason, independent of the state of the connection (and without irrespective of the connection mutex). As a result, the connectino can be in pretty much any state at the time its socket is closed. Handle those other cases at the top of con_work(). Pull this whole block of code into a separate function to reduce the clutter. Signed-off-by:
-
Sage Weil authored
If we are creating an osd request and get an invalid layout, return an EINVAL to the caller. We switch up the return to have an error code instead of NULL implying -ENOMEM. Signed-off-by:
-
Alex Elder authored
If a read-only rbd device is opened for writing in rbd_open(), it returns without dropping the just-acquired device reference. Fix this by moving the read-only check before getting the reference. Signed-off-by:
-
Sage Weil authored
If the MDS gives us a dentry and we weren't prepared to handle it, WARN_ON_ONCE instead of crashing. Reported-by:
Yan, Zheng <zheng.z.yan@intel.com> Signed-off-by:
-
Josh Durgin authored
Previously the original header version was sent. Now, we update it when the header changes. Signed-off-by:
Josh Durgin <josh.durgin@dreamhost.com> Reviewed-by:
-
Josh Durgin authored
This prevents a race between requests with a given snap context and header updates that free it. The osd client was already expecting the snap context to be reference counted, since it get()s it in ceph_osdc_build_request and put()s it when the request completes. Also remove the second down_read()/up_read() on header_rwsem in rbd_do_request, which wasn't actually preventing this race or protecting any other data. Signed-off-by:
-
Josh Durgin authored
The image may have been resized. Signed-off-by:
-
Josh Durgin authored
If an image was mapped to a snapshot, the size of the head version would be shown. Protect capacity with header_rwsem, since it may change. Signed-off-by:
-
Josh Durgin authored
Snapshots cannot be resized, and the new capacity of head should not be reflected by the snapshot. Signed-off-by:
-
Josh Durgin authored
When a snapshot is deleted, the OSD will return ENOENT when reading from it. This is normally interpreted as a hole by rbd, which will return zeroes. To minimize the time in which this can happen, stop requests early when we are notified that our snapshot no longer exists. [elder@inktank.com: updated __rbd_init_snaps_header() logic] Signed-off-by:
-
Sage Weil authored
When we detect a mds session reset, close the old ceph_connection before reopening it. This ensures we clean up the old socket properly and keep the ceph_connection state correct. Signed-off-by:
-
Daniel Vetter authored
commit 48e85834 upstream. This reverts commit 9756fe38. The bogus lvds output is actually a lvds->hdmi bridge, which we don't really support. But unconditionally disabling it breaks some existing setups. Reported-by:
John Tapsell <johnflux@gmail.com> References: http://permalink.gmane.org/gmane.comp.freedesktop.xorg.drivers.intel/17237Signed-off-by:
Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Luis Henriques <luis.henriques@canonical.com> Signed-off-by:
-
Chris Wilson authored
commit 5bb61643 upstream. This was meant to be the purpose of the intel_crtc_wait_for_pending_flips() function which is called whilst preparing the CRTC for a modeset or before disabling. However, as Ville Syrjala pointed out, we set the pending flip notification on the old framebuffer that is no longer attached to the CRTC by the time we come to flush the pending operations. Instead, we can simply wait on the pending unpin work to be finished on this CRTC, knowning that the hardware has therefore finished modifying the registers, before proceeding with our direct access. Fixes i-g-t/flip_test on non-pch platforms. pch platforms simply schedule the flip immediately when the pipe is disabled, leading to other funny issues. Signed-off-by:
Chris Wilson <chris@chris-wilson.co.uk> [danvet: Added i-g-t note and cc: stable] Signed-off-by:
Julien Cristau <jcristau@debian.org> Signed-off-by:
-
Daniel Vetter authored
commit 74d44445 upstream. ... since finish_page_flip needs the vblank timestamp generated in drm_handle_vblank. Somehow all the gmch platforms get it right, but all the pch platform irq handlers get is wrong. Hooray for copy& pasting! Currently this gets papered over by a gross hack in finish_page_flip. A second patch will remove that. Note that without this, the new timestamp sanity checks in flip_test occasionally get tripped up, hence the cc: stable tag. Reviewed-by: mario.kleiner@tuebingen.mpg.de Tested-by:
Imre Deak <imre.deak@intel.com> Signed-off-by:
Ben Hutchings <ben@decadent.org.uk> Signed-off-by:
-
Ben Widawsky authored
commit f8f2ac9a upstream. I can't even find how I figured this might be needed anymore. But sure enough, the value I'm reading back on platforms doesn't match what the docs recommends. It seemed to fix Chris' GT1 in limited testing as well. Tested-by:
Ben Widawsky <ben@bwidawsk.net> Signed-off-by:
-
Eugeni Dodonov authored
commit ab3951eb upstream. We should not hit this under any sane conditions, but still, this does not looks right. Reported-by:
Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> CC: Chris Wilson <chris@chris-wilson.co.uk> CC: Daniel Vetter <daniel.vetter@ffwll.ch> Reviewed-by:
Eugeni Dodonov <eugeni.dodonov@intel.com> Signed-off-by:
-
Chris Wilson authored
commit 7e81a42e upstream. Pin-leaks persist and we get the perennial bug reports of machine lockups to the BUG_ON(pin_count==MAX). If we instead loudly report that the object cannot be pinned at that time it should prevent the driver from locking up, and hopefully restore a semblance of working whilst still leaving us a OOPS to debug. Signed-off-by:
-
Jani Nikula authored
commit fcbc50da upstream. Avoid constant wakeups caused by noisy irq lines when we don't even care about the irq. This should be particularly useful for i945g/gm where the hotplug has been disabled: commit 768b107e Author: Daniel Vetter <daniel.vetter@ffwll.ch> Date: Fri May 4 11:29:56 2012 +0200 drm/i915: disable sdvo hotplug on i945g/gm v2: While at it, remove the bogus hotplug_active read, and do not mask hotplug_active[0] before checking whether the irq is needed, per discussion with Daniel on IRC. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=38442Tested-by:
Dominik Köppl <dominik@devwork.org> Signed-off-by:
Jani Nikula <jani.nikula@intel.com> Signed-off-by:
-
Alex Deucher authored
commit ecd67955 upstream. No functional change, but re-order the cases so they evaluate properly due to the way the DCE macros work. Noticed by kallisti5 on IRC. Signed-off-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
-
Chris Wilson authored
commit 084b612e upstream. Note that gen3 is the only platform where we've got the bit definitions right, hence the workaround of disabling sdvo hotplug support on i945g/gm is not due to misdiagnosis of broken hotplug irq handling ... Signed-off-by:
-
Paulo Zanoni authored
commit 9d9740f0 upstream. On IVB and older, we basically have two registers: the control and the data register. We write a few consecutitve times to the control register, and we need these writes to arrive exactly in the specified order. Also, when we're changing the data register, we need to guarantee that anything written to the control register already arrived (since changing the control register can change where the data register points to). Also, we need to make sure all the writes to the data register happen exactly in the specified order, and we also *can't* read the data register during this process, since reading and/or writing it will change the place it points to. So invoke the "better safe than sorry" rule and just be careful and put barriers everywhere :) On HSW we still have a control register that we write many times, but we have many data registers. Demanded-by:
Paulo Zanoni <paulo.r.zanoni@intel.com> Signed-off-by:
-
Alex Deucher authored
commit 26fe45a0 upstream. Selecting ATOM_PPLL_INVALID should be equivalent as the DCPLL or PPLL0 are already programmed for the DISPCLK, but the preferred method is to always specify the PLL selected. SetPixelClock will check the parameters and skip the programming if the PLL is already set up. Signed-off-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
-
Daniel Vetter authored
commit c07496fa upstream. ... we will botch up the bit17 swizzling. Furthermore tiled pwrite is a (now) unused slowpath, so no one really cares. This fixes the last swizzling issues I have with i-g-t on my bit17 swizzling i915G. No regression, it's been broken since the dawn of gem, but it's nice for regression tracking when really _all_ i-g-t tests work. Actually this is not true, Chris Wilson noticed while reviewing this patch that the commit commit d9e86c0e Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Nov 10 16:40:20 2010 +0000 drm/i915: Pipelined fencing [infrastructure] contained a functional change that broke things. Reviewed-by:
-
Chris Wilson authored
commit 0f91128d upstream. During modeset we have to disable the pipe to reconfigure its timings and maybe its size. Userspace may have queued up command buffers that depend upon the pipe running in a certain configuration and so the commands may become confused across the modeset. At the moment, we use a less than satisfactory kick-scanline-waits should the GPU hang during the modeset. It should be more reliable to wait for the pending operations to complete first, even though we still have a window for userspace to submit a broken command buffer during the modeset. Signed-off-by:
-
Sean Paul authored
commit f01db988 upstream. I have seen a number of "blt ring initialization failed" messages where the ctl or start registers are not the correct value. Upon further inspection, if the code just waited a little bit, it would read the correct value. Adding the wait_for to these reads should eliminate the issue. Signed-off-by:
Sean Paul <seanpaul@chromium.org> Reviewed-by:
-
Takashi Iwai authored
commit b0354385 upstream. Currently i915 driver checks [PCH_]LVDS register bits to decide whether to set up the dual-link or the single-link mode. This relies implicitly on that BIOS initializes the register properly at boot. However, BIOS doesn't initialize it always. When the machine is booted with the closed lid, BIOS skips the LVDS reg initialization. This ends up in blank output on a machine with a dual-link LVDS when you open the lid after the boot. This patch adds a workaround for that problem by checking the initial LVDS register value in VBT. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=37742Tested-By:
Rodrigo Vivi <rodrigo.vivi@gmail.com> Reviewed-by:
Adam Jackson <ajax@redhat.com> Signed-off-by:
Takashi Iwai <tiwai@suse.de> Signed-off-by:
-
Daniel Vetter authored
commit b7884eb4 upstream. Empirical evidence suggests that we need to: On at least one ivb machine when running the hangman i-g-t test, the rings don't properly initialize properly - the RING_START registers seems to be stuck at all zeros. Holding forcewake around this register init sequences makes chip reset reliable again. Note that this is not the first such issue: commit f01db988 Author: Sean Paul <seanpaul@chromium.org> Date: Fri Mar 16 12:43:22 2012 -0400 drm/i915: Add wait_for in init_ring_common added delay loops to make RING_START and RING_CTL initialization reliable on the blt ring at boot-up. So I guess it won't hurt if we do this unconditionally for all force_wake needing gpus. To avoid copy&pasting of the HAS_FORCE_WAKE check I've added a new intel_info bit for that. v2: Fixup missing commas in static struct and properly handling the error case in init_ring_common, both noticed by Jani Nikula. Reported-and-tested-by:
Yang Guang <guang.a.yang@intel.com> Reviewed-by:
-
