- 06 Jan, 2010 32 commits
-
-
Ingo Molnar authored
commit a8fe9ea2 upstream. Stephen Rothwell reported the following build warning: lib/dma-debug.c: In function 'dma_debug_device_change': lib/dma-debug.c:680: warning: 'return' with no value, in function returning non-void Introduced by commit f797d988 ("dma-debug: Do not add notifier when dma debugging is disabled"). Return 0 [notify-done] when disabled. (this is standard bus notifier behavior.) Signed-off-by:
Shaun Ruffell <sruffell@digium.com> Signed-off-by:
Joerg Roedel <joerg.roedel@amd.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> LKML-Reference: <20091231125624.GA14666@liondog.tnic> Signed-off-by:
Ingo Molnar <mingo@elte.hu> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Shaun Ruffell authored
commit f797d988 upstream. If CONFIG_HAVE_DMA_API_DEBUG is defined and "dma_debug=off" is specified on the kernel command line, when you detach a driver from a device you can cause the following NULL pointer dereference: BUG: unable to handle kernel NULL pointer dereference at (null) IP: [<c0580d35>] dma_debug_device_change+0x5d/0x117 The problem is that the dma_debug_device_change notifier function is added to the bus notifier chain even though the dma_entry_hash array was never initialized. If dma debugging is disabled, this patch both prevents dma_debug_device_change notifiers from being added to the chain, and additionally ensures that the dma_debug_device_change notifier function is a no-op. Signed-off-by:
-
Nicolas Ferre authored
commit 4297a462 upstream. Correct a typo error in locking calls. Signed-off-by:
Nicolas Ferre <nicolas.ferre@atmel.com> Signed-off-by:
Dan Williams <dan.j.williams@intel.com> Signed-off-by:
-
NeilBrown authored
commit cbd19983 upstream. evms configures md arrays by: open device send ioctl close device for each different ioctl needed. Since 2.6.29, the device can disappear after the 'close' unless a significant configuration has happened to the device. The change made by "SET_ARRAY_INFO" can too minor to stop the device from disappearing, but important enough that losing the change is bad. So: make sure SET_ARRAY_INFO sets mddev->ctime, and keep the device active as long as ctime is non-zero (it gets zeroed with lots of other things when the array is stopped). This is suitable for -stable kernels since 2.6.29. Signed-off-by:
NeilBrown <neilb@suse.de> Signed-off-by:
-
Mike Travis authored
commit 39d30770 upstream. The wrong address was being used to write the SCIR led regs on remote hubs. Also, there was an inconsistency between how BIOS and the kernel indexed these regs. Standardize on using the lower 6 bits of the APIC ID as the index. This patch fixes the problem of writing to an errant address to a cpu # >= 64. Signed-off-by:
Mike Travis <travis@sgi.com> Reviewed-by:
Jack Steiner <steiner@sgi.com> Cc: Robin Holt <holt@sgi.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> LKML-Reference: <4B3922F9.3060905@sgi.com> [ v2: fix a number of annoying checkpatch artifacts and whitespace noise ] Signed-off-by:
-
Julia Lawall authored
commit 6057912d upstream. sizeof(dev->dev_addr) is the size of a pointer. A few lines above, the size of this field is obtained using netdev->addr_len for a call to memcpy, so do the same here. A simplified version of the semantic patch that finds this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ expression *x; expression f; type T; @@ *f(...,(T)x,...) // </smpl> Signed-off-by:
Julia Lawall <julia@diku.dk> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
Alan Stern authored
commit da307123 upstream. This patch (as1315) fixes some bugs in the USB core authorization code: usb_deauthorize_device() should deallocate the device strings instead of leaking them, and it should invoke usb_destroy_configuration() (which does proper reference counting) instead of freeing the config information directly. usb_authorize_device() shouldn't change the device strings until it knows that the authorization will succeed, and it should autosuspend the device at the end (having autoresumed the device at the start). Because the device strings can be changed, the sysfs routines to display the strings must protect the string pointers by locking the device. Signed-off-by:
Alan Stern <stern@rowland.harvard.edu> CC: Inaky Perez-Gonzalez <inaky@linux.intel.com> Acked-by:
David Vrabel <david.vrabel@csr.com> Signed-off-by:
-
Alan Stern authored
commit 8d8558d1 upstream. This patch (as1314) renames usb_configure_device() and usb_configure_device_otg() in the hub driver. Neither name is appropriate because these routines enumerate devices, they don't configure them. That's handled by usb_choose_configuration() and usb_set_configuration(). Signed-off-by:
-
Oliver Neukum authored
commit 652fd781 upstream. The device must be marked busy as it receives data. Signed-off-by:
Oliver Neukum <oliver@neukum.org> Tested-by:
Matthew Garrett <mjg@redhat.com> Signed-off-by:
Marcel Holtmann <marcel@holtmann.org> Signed-off-by:
-
Sergei Shtylyov authored
commit 17be5c5f upstream. Gadget stalling a zero-length SETUP request results in this error message: SetupEnd came in a wrong ep0stage idle In order to avoid it, always set the CSR0.DataEnd bit after detecting a zero- length request. Add the missing '\n' to the error message itself as well... Signed-off-by:
Sergei Shtylyov <sshtylyov@ru.mvista.com> Acked-by:
Anand Gadiyar <gadiyar@ti.com> Signed-off-by:
Felipe Balbi <felipe.balbi@nokia.com> Signed-off-by:
-
pancho horrillo authored
commit 37e9066b upstream. brightness status is reported by the Apple Cinema Displays as an 'unsigned char' (u8) value, but the code used 'char' instead. Note that he driver was developed on the PowerPC architecture, where the two types are synonymous, which is not always the case. Fixed that. Otherwise the driver will interpret brightness levels > 127 as negative, and fail to load. Signed-off-by:
pancho horrillo <pancho@pancho.name> Signed-off-by:
-
Donny Kurnia authored
commit c983202b upstream. I made this patch for usbserial driver to add the support for EVDO modem Haier CE100. The bugs report for this is here: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/490068 This patch based on these post: http://blankblondtank.wordpress.com/2009/09/04/mengoptimalkan-koneksi-modem-haier-ce-100-cdma-di-linux/ http://tantos.web.id/blogs/how-to-internet-connection-using-cdma-evdo-modem-and-karmic-koala-ubuntu-9-10 I hope this patch can help other that have the Haier C100 modem, mostly in my country, Indonesia. Signed-off-by:
Donny Kurnia <donnykurnia@gmail.com> Signed-off-by:
-
Clemens Ladisch authored
commit ac06c067 upstream. While converting emi62 to use request_firmware(), the driver was also changed to use the ihex helper functions. However, this broke the loading of the FPGA firmware because the code tries to access the addr field of the EOF record which works with a plain array that has an empty last record but not with the ihex helper functions where the end of the data is signaled with a NULL record pointer, resulting in: BUG: unable to handle kernel NULL pointer dereference at (null) IP: [<f80d248c>] emi62_load_firmware+0x33c/0x740 [emi62] This can be fixed by changing the loop condition to test the return value of ihex_next_binrec() directly (like in emi26.c). Signed-off-by:
Clemens Ladisch <clemens@ladisch.de> Reported-and-tested-by:
Der Mickster <retroeffective@gmail.com> Acked-by:
David Woodhouse <David.Woodhouse@intel.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Dave Airlie authored
commit 794f3141 upstream. drivers/gpu/drm/radeon/radeon_test.c:45: undefined reference to `__udivdi3' Reported-by:
Mr. James W. Laferriere <babydr@baby-dragons.com> Signed-off-by:
Dave Airlie <airlied@redhat.com> Signed-off-by:
-
Eric Millbrandt authored
commit 48e3cbb3 upstream. This patch fixes a bug where "virtual" registers were being written to the ac97 bus. This was causing unrelated registers to become corrupted (headphone 0x04, touchscreen 0x78, etc). This patch duplicates protection that was included in the wm9713 driver. Signed-off-by:
Eric Millbrandt <emillbrandt@dekaresearch.com> Acked-by:
Liam Girdwood <lrg@slimlogic.co.uk> Signed-off-by:
Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by:
-
Neil Campbell authored
commit bb7f20b1 upstream. This patch fixes the handling of VSX alignment faults in little-endian mode (the current code assumes the processor is in big-endian mode). The patch also makes the handlers clear the top 8 bytes of the register when handling an 8 byte VSX load. This is based on 2.6.32. Signed-off-by:
Neil Campbell <neilc@linux.vnet.ibm.com> Acked-by:
Michael Neuling <mikey@neuling.org> Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Zhao Yakui authored
commit 13c199c0 upstream. On some laptops it will return NOTIFY_OK(non-zero) when calling the ACPI LID notifier. Then it is used as the result of ACPI LID resume function, which will complain the following warning message in course of suspend/resume: >PM: Device PNP0C0D:00 failed to resume: error 1 This patch is to eliminate the above warning message. http://bugzilla.kernel.org/show_bug.cgi?id=14782Signed-off-by:
Zhao Yakui <yakui.zhao@intel.com> Signed-off-by:
Len Brown <len.brown@intel.com> Signed-off-by:
-
Alexey Starikovskiy authored
commit 55b313f2 upstream. MSI strings should be ORed, not ANDed. Reference: http://bugzilla.kernel.org/show_bug.cgi?id=14446Signed-off-by:
Alexey Starikovskiy <astarikovskiy@suse.de> Signed-off-by:
-
Stefan Bader authored
commit bdc731bc upstream. BugLink: https://bugs.launchpad.net/ubuntu/+bug/435958 The module alias currently matches any Acer computer but when loaded the BIOS checks will only succeed on Aspire One models. This causes a invalid BIOS warning for all other models (seen on Aspire 4810T). This is not fatal but worries users that see this message. Limiting the moule alias to models starting with AOA or DOA for Packard Bell. Signed-off-by:
Stefan Bader <stefan.bader@canonical.com> Acked-by:
Borislav Petkov <petkovbb@gmail.com> Acked-by:
Peter Feuerer <peter@piie.net> Signed-off-by:
-
Takashi Iwai authored
commit 035eb0cf upstream. Some model quirks missed the corresponding capsrc_nids. This resulted in non-working capture source selection. Signed-off-by:
Takashi Iwai <tiwai@suse.de> Signed-off-by:
-
Clemens Ladisch authored
commit 3e85fd61 upstream. When allocating the PCM buffer, use vmalloc_user() instead of vmalloc(). Otherwise, it would be possible for applications to play the previous contents of the kernel memory to the speakers, or to read it directly if the buffer is exported to userspace. Signed-off-by:
-
Guennadi Liakhovetski authored
commit 48c03ce7 upstream. The wm8974 datasheet defines BUFIOEN as bit 2. Signed-off-by:
Guennadi Liakhovetski <g.liakhovetski@gmx.de> Acked-by:
-
Bartlomiej Zolnierkiewicz authored
commit 509426bd upstream. adev->dma_mode stores the transfer mode value not UDMA mode number so the condition in cmd64x_set_dmamode() is always true and the higher UDMA clock is always selected. This can potentially result in data corruption when UDMA33 device is used, when 40-wire cable is used or when the error recovery code decides to lower the device speed down. The issue was introduced in the commit 6a40da02 ("libata cmd64x: whack into a shape that looks like the documentation") which goes back to kernel 2.6.20. Signed-off-by:
Bartlomiej Zolnierkiewicz <bzolnier@gmail.com> Signed-off-by:
Jeff Garzik <jgarzik@redhat.com> Signed-off-by:
-
Sergei Shtylyov authored
commit 256ace9b upstream. The clock turnaround code still doesn't work for several reasons: - 'USE_DPLL' flag in 'ap->host->private_data' is never initialized or updated, so the driver can only set the chip to the DPLL clock mode, not the PCI mode; - the driver doesn't serialize access to the channels depending on the current clock mode like the vendor drivers, so the clock turnaround is only executed "optionally", not always as it should be; - the wrong ports are written to when hpt3x2n_set_clock() is called for the secondary channel; - hpt3x2n_set_clock() can inadvertently enable the disabled channels when resetting the channel state machines. Signed-off-by:
-
Thomas Gleixner authored
commit bb6eddf7 upstream. Xiaotian Feng triggered a list corruption in the clock events list on CPU hotplug and debugged the root cause. If a CPU registers more than one per cpu clock event device, then only the active clock event device is removed on CPU_DEAD. The unused devices are kept in the clock events device list. On CPU up the clock event devices are registered again, which means that we list_add an already enqueued list_head. That results in list corruption. Resolve this by removing all devices which are associated to the dead CPU on CPU_DEAD. Reported-by:
Xiaotian Feng <dfeng@redhat.com> Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Tested-by:
-
Peter Zijlstra authored
commit e4f42888 upstream. We should skip !SD_LOAD_BALANCE domains. Signed-off-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <20091216170517.653578430@chello.nl> Signed-off-by:
-
Suresh Siddha authored
commit 45a94d7c upstream. xsave_cntxt_init() does something like: cpuid(0xd, ..); // find out what features FP/SSE/.. etc are supported xsetbv(); // enable the features known to OS cpuid(0xd, ..); // find out the size of the context for features enabled Depending on what features get enabled in xsetbv(), value of the cpuid.eax=0xd.ecx=0.ebx changes correspondingly (representing the size of the context that is enabled). As we don't have volatile keyword for native_cpuid(), gcc 4.1.2 optimizes away the second cpuid and the kernel continues to use the cpuid information obtained before xsetbv(), ultimately leading to kernel crash on processors supporting more state than the legacy FP/SSE. Add "volatile" for native_cpuid(). Signed-off-by:
Suresh Siddha <suresh.b.siddha@intel.com> LKML-Reference: <1261009542.2745.55.camel@sbs-t61.sc.intel.com> Signed-off-by:
H. Peter Anvin <hpa@zytor.com> Signed-off-by:
-
Peter Zijlstra authored
commit e6c8fba7 upstream. Make sure not to access sched_fair fields before verifying it is indeed a sched_fair task. Signed-off-by:
-
Mike Christie authored
commit 48de68a4 upstream. If transport_class_register fails we should unregister any registered classes, or we will leak memory or other resources. I did a quick modprobe of scsi_transport_fc to test the patch. Signed-off-by:
Mike Christie <michaelc@cs.wisc.edu> Signed-off-by:
James Bottomley <James.Bottomley@suse.de> Signed-off-by:
-
FUJITA Tomonori authored
commit c982c368 upstream. dio transfer always resets mdata->page_order to zero. It breaks high-order pages previously allocated for non-dio transfer. This patches adds reserved_page_order to st_buffer structure to save page order for non-dio transfer. http://bugzilla.kernel.org/show_bug.cgi?id=14563 When enlarge_buffer() allocates 524288 from 0, st uses six-order page allocation. So mdata->page_order is 6 and frp_seg is 2. After that, if st uses dio, sgl_map_user_pages() sets mdata->page_order to 0 for st_do_scsi(). After that, when we call normalize_buffer(), it frees only free frp_seg * PAGE_SIZE (2 * 4096) though we should free frp_seg * PAGE_SIZE << 6 (2 * 4096 << 6). So we see buffer_size is set to 516096 (524288 - 8192). Reported-by:
Joachim Breuer <linux-kernel@jmbreuer.net> Tested-by:
Kai Makisara <kai.makisara@kolumbus.fi> Signed-off-by:
FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by:
-
Michael Reed authored
commit 1486400f upstream. Fix crash in qla2x00_fdmi_register() due to the dpc thread executing before the scsi host has been fully added. Unable to handle kernel NULL pointer dereference (address 00000000000001d0) qla2xxx_7_dpc[4140]: Oops 8813272891392 [1] Call Trace: [<a000000100016910>] show_stack+0x50/0xa0 sp=e00000b07c59f930 bsp=e00000b07c591400 [<a000000100017180>] show_regs+0x820/0x860 sp=e00000b07c59fb00 bsp=e00000b07c5913a0 [<a00000010003bd60>] die+0x1a0/0x2e0 sp=e00000b07c59fb00 bsp=e00000b07c591360 [<a0000001000681a0>] ia64_do_page_fault+0x8c0/0x9e0 sp=e00000b07c59fb00 bsp=e00000b07c591310 [<a00000010000c8e0>] ia64_native_leave_kernel+0x0/0x270 sp=e00000b07c59fb90 bsp=e00000b07c591310 [<a000000207197350>] qla2x00_fdmi_register+0x850/0xbe0 [qla2xxx] sp=e00000b07c59fd60 bsp=e00000b07c591290 [<a000000207171570>] qla2x00_configure_loop+0x1930/0x34c0 [qla2xxx] sp=e00000b07c59fd60 bsp=e00000b07c591128 [<a0000002071732b0>] qla2x00_loop_resync+0x1b0/0x2e0 [qla2xxx] sp=e00000b07c59fdf0 bsp=e00000b07c5910c0 [<a000000207166d40>] qla2x00_do_dpc+0x9a0/0xce0 [qla2xxx] sp=e00000b07c59fdf0 bsp=e00000b07c590fa0 [<a0000001000d5bb0>] kthread+0x110/0x140 sp=e00000b07c59fe00 bsp=e00000b07c590f68 [<a000000100014a30>] kernel_thread_helper+0xd0/0x100 sp=e00000b07c59fe30 bsp=e00000b07c590f40 [<a00000010000a4c0>] start_kernel_thread+0x20/0x40 sp=e00000b07c59fe30 bsp=e00000b07c590f40 crash> dis a000000207197350 0xa000000207197350 <qla2x00_fdmi_register+2128>: [MMI] ld1 r45=[r14];; crash> scsi_qla_host.host 0xe00000b058c73ff8 host = 0xe00000b058c73be0, crash> Scsi_Host.shost_data 0xe00000b058c73be0 shost_data = 0x0, <<<<<<<<<<< The fc_transport fc_* workqueue threads have yet to be created. crash> ps | grep _7 3891 2 2 e00000b075c80000 IN 0.0 0 0 [scsi_eh_7] 4140 2 3 e00000b07c590000 RU 0.0 0 0 [qla2xxx_7_dpc] The thread creating adding the Scsi_Host is blocked due to other activity in sysfs. crash> bt 3762 PID: 3762 TASK: e00000b071e70000 CPU: 3 COMMAND: "modprobe" #0 [BSP:e00000b071e71548] schedule at a000000100727e00 #1 [BSP:e00000b071e714c8] __mutex_lock_slowpath at a0000001007295a0 #2 [BSP:e00000b071e714a8] mutex_lock at a000000100729830 #3 [BSP:e00000b071e71478] sysfs_addrm_start at a0000001002584f0 #4 [BSP:e00000b071e71440] create_dir at a000000100259350 #5 [BSP:e00000b071e71410] sysfs_create_subdir at a000000100259510 #6 [BSP:e00000b071e713b0] internal_create_group at a00000010025c880 #7 [BSP:e00000b071e71388] sysfs_create_group at a00000010025cc50 #8 [BSP:e00000b071e71368] dpm_sysfs_add at a000000100425050 #9 [BSP:e00000b071e71310] device_add at a000000100417d90 #10 [BSP:e00000b071e712d8] scsi_add_host at a00000010045a380 #11 [BSP:e00000b071e71268] qla2x00_probe_one at a0000002071be950 #12 [BSP:e00000b071e71248] local_pci_probe at a00000010032e490 #13 [BSP:e00000b071e71218] pci_device_probe at a00000010032ecd0 #14 [BSP:e00000b071e711d8] driver_probe_device at a00000010041d480 #15 [BSP:e00000b071e711a8] __driver_attach at a00000010041d6e0 #16 [BSP:e00000b071e71170] bus_for_each_dev at a00000010041c240 #17 [BSP:e00000b071e71150] driver_attach at a00000010041d0a0 #18 [BSP:e00000b071e71108] bus_add_driver at a00000010041b080 #19 [BSP:e00000b071e710c0] driver_register at a00000010041dea0 #20 [BSP:e00000b071e71088] __pci_register_driver at a00000010032f610 #21 [BSP:e00000b071e71058] (unknown) at a000000207200270 #22 [BSP:e00000b071e71018] do_one_initcall at a00000010000a9c0 #23 [BSP:e00000b071e70f98] sys_init_module at a0000001000fef00 #24 [BSP:e00000b071e70f98] ia64_ret_from_syscall at a00000010000c740 So, it appears that qla2xxx dpc thread is moving forward before the scsi host has been completely added. This patch moves the setting of the init_done (and online) flag to after the call to scsi_add_host() to hold off the dpc thread. Found via large lun count testing using 2.6.31. Signed-off-by:
Michael Reed <mdr@sgi.com> Acked-by:
Giridhar Malavali <giridhar.malavali@qlogic.com> Signed-off-by:
-
Kleber Sacilotto de Souza authored
commit 99c965dd upstream. After commits c82f63e4 (PCI: check saved state before restore) and 4b77b0a2 (PCI: Clear saved_state after the state has been restored) PCI drivers are prevented from restoring the device standard configuration registers twice in a row. These changes introduced a regression on ipr EEH recovery. The ipr device driver saves the PCI state only during the device probe and restores it on ipr_reset_restore_cfg_space() during IOA resets. This behavior is causing the EEH recovery to fail after the second error detected, since the registers are not being restored. One possible solution would be saving the registers after restoring them. The problem with this approach is that while recovering from an EEH error if pci_save_state() results in an EEH error, the adapter/slot will be reset, and end up back in ipr_reset_restore_cfg_space(), but it won't have a valid saved state to restore, so pci_restore_state() will fail. The following patch introduces a workaround for this problem, hacking around the PCI API by setting pdev->state_saved = true before we do the restore. It fixes the EEH regression and prevents that we hit another EEH error during EEH recovery. [jejb: fix is a hack ... Jesse and Rafael will fix properly] Signed-off-by:
Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by:
Brian King <brking@linux.vnet.ibm.com> Cc: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by:
-
- 18 Dec, 2009 8 commits
-
-
Greg Kroah-Hartman authored
-
Luck, Tony authored
commit cd7bcf32 upstream. Needed for commit 2c992208 ("intel-iommu: Detect DMAR in hyperspace at probe time.) to build on IA64. Signed-off-by:
Tony Luck <tony.luck@intel.com> Signed-off-by:
-
Paul Mackerras authored
commit 0f624e7e upstream. It is quite legitimate for CPUs to be numbered sparsely, meaning that it possible for an online CPU to have a number which is greater than the total count of possible CPUs. Currently find_get_context() has a sanity check on the cpu number where it checks it against num_possible_cpus(). This test can fail for a legitimate cpu number if the cpu_possible_mask is sparsely populated. This fixes the problem by checking the CPU number against nr_cpumask_bits instead, since that is the appropriate check to ensure that the cpu number is same to pass to cpu_isset() subsequently. Reported-by:
Paul Mackerras <paulus@samba.org> Tested-by:
Peter Zijlstra <peterz@infradead.org> LKML-Reference: <20091215084032.GA18661@brick.ozlabs.ibm.com> Signed-off-by:
-
Jan Engelhardt authored
commit 7a922637 upstream. For both .33 and .32-stable. Signed-off-by:
Jan Engelhardt <jengelh@medozas.de> Signed-off-by:
Patrick McHardy <kaber@trash.net> Signed-off-by:
-
Chris Wright authored
commit 1672af11 upstream. We are seeing a bug when booting w/ iommu=pt with current upstream (bisect blames 19943b0e "intel-iommu: Unify hardware and software passthrough support). The issue is specific to this loop during identity map initialization of each device: domain_context_mapping_one(si_domain, ..., CONTEXT_TT_PASS_THROUGH) ... /* Skip top levels of page tables for * iommu which has less agaw than default. */ for (agaw = domain->agaw; agaw != iommu->agaw; agaw--) { pgd = phys_to_virt(dma_pte_addr(pgd)); if (!dma_pte_present(pgd)) { <------ failing here spin_unlock_irqrestore(&iommu->lock, flags); return -ENOMEM; } This box has 2 iommu's in it. The catchall iommu has MGAW == 48, and SAGAW == 4. The other iommu has MGAW == 39, SAGAW == 2. The device that's failing the above pgd test is the only device connected to the non-catchall iommu, which has a smaller address width than the domain default. This test is not necessary since the context is in PT mode and the ASR is ignored. Thanks to Don Dutile for discovering and debugging this one. Signed-off-by:
Chris Wright <chrisw@sous-sol.org> Signed-off-by:
-
David Woodhouse authored
commit 44cd613c upstream. The hotplug notifier will call find_domain() to see if the device in question has been assigned an IOMMU domain. However, this should never be called for devices with a "dummy" domain, such as graphics devices when intel_iommu=igfx_off is set and the corresponding IOMMU isn't even initialised. If you do that, it'll oops as it dereferences the (-1) pointer. The notifier function should check iommu_no_mapping() for the device before doing anything else. Signed-off-by:
-
David Woodhouse authored
commit 5595b528 upstream. Some HP BIOSes report an RMRR region (a region which needs a 1:1 mapping in the IOMMU for a given device) which has an end address lower than its start address. Detect that and warn, rather than triggering the BUG() in dma_pte_clear_range(). Signed-off-by:
-
David Woodhouse authored
commit 6ecbf01c upstream. The BIOS errors where an IOMMU is reported either at zero or a bogus address are causing problems even when the IOMMU is disabled -- because interrupt remapping uses the same hardware. Ensure that the checks get applied for the interrupt remapping initialisation too. Signed-off-by:
-
