aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2018-04-10Merge tag 'fbdev-v4.17' of git://github.com/bzolnier/linuxLinus Torvalds30-818/+609
Pull fbdev updates from Bartlomiej Zolnierkiewicz: "There is nothing really major here, just a couple of small bugfixes, improvements and cleanups: - make it possible to load radeonfb driver when offb driver is loaded first (Mathieu Malaterre) - fix memory leak in offb driver (Mathieu Malaterre) - fix unaligned access in udlfb driver (Ladislav Michl) - convert atmel_lcdfb driver to use GPIO descriptors (Ludovic Desroches) - avoid mismatched prototypes in sisfb driver (Arnd Bergmann) - remove VLA usage from viafb driver (Gustavo A. R. Silva) - add missing help text to FB_I810_I2 config option (Ulf Magnusson) - misc fixes (Gustavo A. R. Silva, Colin Ian King, Markus Elfring) - remove dead code from s3c-fb driver for Exynos and S5PV210 platforms - misc cleanups (Corentin Labbe, Ladislav Michl, Ulf Magnusson, Vladimir Zapolskiy, Markus Elfring)" * tag 'fbdev-v4.17' of git://github.com/bzolnier/linux: (32 commits) video: fbdev: s3c-fb: remove dead platform code for Exynos and S5PV210 platforms video: au1100fb: Delete an unnecessary variable initialisation in au1100fb_drv_probe() video: au1100fb: Improve a size determination in au1100fb_drv_probe() video: au1100fb: Delete an error message for a failed memory allocation in au1100fb_drv_probe() video/console/sticore: Delete an error message for a failed memory allocation in sti_try_rom_generic() video: ARM CLCD: Improve a size determination in clcdfb_probe() video: ARM CLCD: Delete an error message for a failed memory allocation in clcdfb_probe() video: matroxfb: Delete an error message for a failed memory allocation in matroxfb_crtc2_probe() video: s3c-fb: Improve a size determination in s3c_fb_probe() video: s3c-fb: Delete an error message for a failed memory allocation in s3c_fb_probe() video: fsl-diu-fb: Delete an error message for a failed memory allocation in fsl_diu_init() video: ssd1307fb: Improve a size determination in ssd1307fb_probe() video: smscufx: Delete an error message for a failed memory allocation in ufx_realloc_framebuffer() video: smscufx: Return an error code only as a constant in ufx_realloc_framebuffer() video: smscufx: Less checks in ufx_usb_probe() after error detection video: udlfb: Return an error code only as a constant in dlfb_realloc_framebuffer() video/fbdev/stifb: Delete an error message for a failed memory allocation in stifb_init_fb() video/fbdev/stifb: Return -ENOMEM after a failed kzalloc() in stifb_init_fb() video: fbdev: aty128fb: use true and false for boolean values fbdev: aty: fix missing indentation in if statement ...
2018-04-10Merge tag 'sound-fix-4.17-rc1' of ↵Linus Torvalds3-77/+58
git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound Pull sound fixes from Takashi Iwai: "The main purpose of this pull request is a fix for a regression in the recent PCM OSS emulation code that may lead to RCU stall. Since syzkaller hits this too often, I send the pull request now with a minimal collection. Possibly another pull request may follow before RC1. The other fixes here are for USB-audio class 2 and 3 to improve the parser for the clock descriptors. These are rather cleanups but good for security, too. Last but not least, another included fix is the trivial one to remove superfluous WARN_ON() that annoyed syzbot" * tag 'sound-fix-4.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: ALSA: pcm: Remove WARN_ON() at snd_pcm_hw_params() error ALSA: pcm: Fix endless loop for XRUN recovery in OSS emulation ALSA: usb-audio: Add sanity checks in UAC3 clock parsers ALSA: usb-audio: More strict sanity checks for clock parsers ALSA: usb-audio: Refactor clock finder helpers
2018-04-10Merge tag 'media/v4.17-2' of ↵Linus Torvalds111-1249/+815
git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media Pull media fixes from Mauro Carvalho Chehab: "A series of media updates/fixes for 4.17. There are two important core fix patches in this series: - A regression fix on Kernel 4.16 with causes it to not work with some input devices that depend on media core - A fix at compat32 bits with causes it to OOPS on overlay, and affects the Kernels where the CVE-2017-13166 was backported The remaining ones are other random fixes at the documentation and on drivers. The biggest part of this series is a set of 18 patches for the Intel atomisp driver. Currently, it produces hundreds of warnings/errors on sparse/smatch, causing me to sometimes ignore new warnings on other drivers that are not so broken. This driver is on really poor state, even for staging standards: it has several layers of abstraction on it, and it supports two different hardware. Selecting between them require to add a define (there isn't even a Kconfig option for such purpose). Just on this smatch cleanup, I could easily get rid of 8 "do-nothing" files. So, I'm seriously considering its removal from upstream, if I don't see any real work on addressing the problems there along this year" * tag 'media/v4.17-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media: (48 commits) media: v4l2-core: fix size of devnode_nums[] bitarray media: v4l2-compat-ioctl32: don't oops on overlay media: i2c: adv748x: afe: fix sparse warning media: extended-controls.rst: transmitter -> receiver media: staging: atomisp: stop duplicating input format types media: staging: atomisp: get rid of an unused var media: staging: atomisp: stop mixing enum types media: staging: atomisp: get rid of some static warnings media: staging: atomisp: use %p to print pointers media: staging: atomisp: remove an useless check media: staging: atomisp: avoid a warning if 32 bits build media: staging: atomisp: don't access a NULL var media: staging: atomisp: Get rid of *default.host.[ch] media: staging: atomisp: get rid of an unused function media: staging: atomisp: remove unused set_pd_base() media: staging: atomisp: fix endianess issues media: staging: atomisp: add a missing include media: staging: atomisp: get rid of stupid statements media: staging: atomisp: declare static vars as such media: staging: atomisp: ia_css_output.host: don't use var before check ...
2018-04-10xfs: only cancel cow blocks when truncating the data forkDarrick J. Wong1-6/+8
In xfs_itruncate_extents, only cancel cow blocks and clear the reflink flag if we were asked to truncate the data fork. Attr fork blocks cannot be shared, so this makes no sense. Signed-off-by: Darrick J. Wong <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]>
2018-04-10ip_gre: clear feature flags when incompatible o_flags are setSabrina Dubroca1-0/+6
Commit dd9d598c6657 ("ip_gre: add the support for i/o_flags update via netlink") added the ability to change o_flags, but missed that the GSO/LLTX features are disabled by default, and only enabled some gre features are unused. Thus we also need to disable the GSO/LLTX features on the device when the TUNNEL_SEQ or TUNNEL_CSUM flags are set. These two examples should result in the same features being set: ip link add gre_none type gre local 192.168.0.10 remote 192.168.0.20 ttl 255 key 0 ip link set gre_none type gre seq ip link add gre_seq type gre local 192.168.0.10 remote 192.168.0.20 ttl 255 key 1 seq Fixes: dd9d598c6657 ("ip_gre: add the support for i/o_flags update via netlink") Signed-off-by: Sabrina Dubroca <[email protected]> Reviewed-by: Xin Long <[email protected]> Acked-by: William Tu <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2018-04-11Kbuild: fix # escaping in .cmd files for future MakeRasmus Villemoes4-5/+9
I tried building using a freshly built Make (4.2.1-69-g8a731d1), but already the objtool build broke with orc_dump.c: In function ‘orc_dump’: orc_dump.c:106:2: error: ‘elf_getshnum’ is deprecated [-Werror=deprecated-declarations] if (elf_getshdrnum(elf, &nr_sections)) { Turns out that with that new Make, the backslash was not removed, so cpp didn't see a #include directive, grep found nothing, and -DLIBELF_USE_DEPRECATED was wrongly put in CFLAGS. Now, that new Make behaviour is documented in their NEWS file: * WARNING: Backward-incompatibility! Number signs (#) appearing inside a macro reference or function invocation no longer introduce comments and should not be escaped with backslashes: thus a call such as: foo := $(shell echo '#') is legal. Previously the number sign needed to be escaped, for example: foo := $(shell echo '\#') Now this latter will resolve to "\#". If you want to write makefiles portable to both versions, assign the number sign to a variable: C := \# foo := $(shell echo '$C') This was claimed to be fixed in 3.81, but wasn't, for some reason. To detect this change search for 'nocomment' in the .FEATURES variable. This also fixes up the two make-cmd instances to replace # with $(pound) rather than with \#. There might very well be other places that need similar fixup in preparation for whatever future Make release contains the above change, but at least this builds an x86_64 defconfig with the new make. Link: https://bugzilla.kernel.org/show_bug.cgi?id=197847 Cc: Randy Dunlap <[email protected]> Signed-off-by: Rasmus Villemoes <[email protected]> Signed-off-by: Masahiro Yamada <[email protected]>
2018-04-10x86/apic: Fix signedness bug in APIC ID validity checksLi RongQing8-13/+16
The APIC ID as parsed from ACPI MADT is validity checked with the apic->apic_id_valid() callback, which depends on the selected APIC type. For non X2APIC types APIC IDs >= 0xFF are invalid, but values > 0x7FFFFFFF are detected as valid. This happens because the 'apicid' argument of the apic_id_valid() callback is type 'int'. So the resulting comparison apicid < 0xFF evaluates to true for all unsigned int values > 0x7FFFFFFF which are handed to default_apic_id_valid(). As a consequence, invalid APIC IDs in !X2APIC mode are considered valid and accounted as possible CPUs. Change the apicid argument type of the apic_id_valid() callback to u32 so the evaluation is unsigned and returns the correct result. [ tglx: Massaged changelog ] Signed-off-by: Li RongQing <[email protected]> Signed-off-by: Thomas Gleixner <[email protected]> Cc: [email protected] Cc: [email protected] Cc: Dou Liyang <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: [email protected] Link: https://lkml.kernel.org/r/[email protected]
2018-04-10MAINTAINERS: Migrate oxnas list to groups.ioNeil Armstrong1-1/+1
The linux-oxnas migrates from tuxfamily to groups.io for a simpler administration and maintainance. Signed-off-by: Neil Armstrong <[email protected]> Signed-off-by: Arnd Bergmann <[email protected]>
2018-04-10arm: npcm: enable L2 cache in NPCM7xx architectureTomer Maimon1-0/+2
This patch Enable ARM L2 cache module in Nuvoton NPCM7xx BMC by adding L2 cache parameters into NPCM7xx DT machine start structure. At patch V7 arm: npcm: add basic support for Nuvoton BMCs we got comments regarding the flags use in L2 cache module. - https://www.spinics.net/lists/arm-kernel/msg613212.html After checking again the L2 cache use in the NPCM7xx, the only L2 cache flag we need to set is L2C_AUX_CTRL_SHARED_OVERRIDE and it is done in the device tree: https://patchwork.kernel.org/patch/10063497/ L2 cache flag mask allowed all the flag option. Signed-off-by: Tomer Maimon <[email protected]> Signed-off-by: Arnd Bergmann <[email protected]>
2018-04-10backing: silence compiler warning using __printfMathieu Malaterre1-0/+1
__printf marker was added in commit d2cc4dde9206 ("bdi_register: add __printf verification, fix arg mismatch") for function `bdi_register` since it is useful to verify format and arguments. Apply equivalent gcc attribute to `bdi_register_va`. Remove warning triggered with W=1: mm/backing-dev.c:881:2: warning: function might be possible candidate for ‘gnu_printf’ format attribute [-Wsuggest-attribute=format] Reviewed-by: Jan Kara <[email protected]> Signed-off-by: Mathieu Malaterre <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10blk-mq: remove code for dealing with remapping queueMing Lei1-31/+3
Firstly, from commit 4b855ad37194 ("blk-mq: Create hctx for each present CPU), blk-mq doesn't remap queue any more after CPU topo is changed. Secondly, set->nr_hw_queues can't be bigger than nr_cpu_ids, and now we map all possible CPUs to hw queues, so at least one CPU is mapped to each hctx. So queue mapping has became static and fixed just like percpu variable, and we don't need to handle queue remapping any more. Cc: Stefan Haberland <[email protected]> Tested-by: Christian Borntraeger <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Signed-off-by: Ming Lei <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10blk-mq: reimplement blk_mq_hw_queue_mappedMing Lei1-1/+1
Now the actual meaning of queue mapped is that if there is any online CPU mapped to this hctx, so implement blk_mq_hw_queue_mapped() in this way. Cc: Stefan Haberland <[email protected]> Tested-by: Christian Borntraeger <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Signed-off-by: Ming Lei <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10blk-mq: don't check queue mapped in __blk_mq_delay_run_hw_queue()Ming Lei1-3/+0
There are several reasons for removing the check: 1) blk_mq_hw_queue_mapped() returns true always now since each hctx may be mapped by one CPU at least 2) when there isn't any online CPU mapped to this hctx, there won't be any IO queued to this CPU, blk_mq_run_hw_queue() only runs queue if there is IO queued to this hctx 3) If __blk_mq_delay_run_hw_queue() is called by blk_mq_delay_run_hw_queue(), which is run from blk_mq_dispatch_rq_list() or scsi_mq_get_budget(), and the hctx to be handled has to be mapped. Cc: Stefan Haberland <[email protected]> Tested-by: Christian Borntraeger <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Signed-off-by: Ming Lei <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10blk-mq: remove blk_mq_delay_queue()Ming Lei3-31/+2
No driver uses this interface any more, so remove it. Cc: Stefan Haberland <[email protected]> Tested-by: Christian Borntraeger <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Signed-off-by: Ming Lei <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10blk-mq: introduce blk_mq_hw_queue_first_cpu() to figure out first cpuMing Lei1-12/+11
This patch introduces helper of blk_mq_hw_queue_first_cpu() for figuring out the hctx's first cpu, and code duplication can be avoided. Cc: Stefan Haberland <[email protected]> Tested-by: Christian Borntraeger <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Signed-off-by: Ming Lei <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10blk-mq: avoid to write intermediate result to hctx->next_cpuMing Lei1-8/+9
This patch figures out the final selected CPU, then writes it to hctx->next_cpu once, then we can avoid to intermediate next cpu observed from other dispatch paths. Cc: Stefan Haberland <[email protected]> Tested-by: Christian Borntraeger <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Signed-off-by: Ming Lei <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10blk-mq: don't keep offline CPUs mapped to hctx 0Ming Lei1-5/+0
From commit 4b855ad37194 ("blk-mq: Create hctx for each present CPU), blk-mq doesn't remap queue after CPU topo is changed, that said when some of these offline CPUs become online, they are still mapped to hctx 0, then hctx 0 may become the bottleneck of IO dispatch and completion. This patch sets up the mapping from the beginning, and aligns to queue mapping for PCI device (blk_mq_pci_map_queues()). Cc: Stefan Haberland <[email protected]> Cc: Keith Busch <[email protected]> Cc: [email protected] Fixes: 4b855ad37194 ("blk-mq: Create hctx for each present CPU) Tested-by: Christian Borntraeger <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Signed-off-by: Ming Lei <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10blk-mq: make sure that correct hctx->next_cpu is setMing Lei1-0/+2
From commit 20e4d81393196 (blk-mq: simplify queue mapping & schedule with each possisble CPU), one hctx can be mapped from all offline CPUs, then hctx->next_cpu can be set as wrong. This patch fixes this issue by making hctx->next_cpu pointing to the first CPU in hctx->cpumask if all CPUs in hctx->cpumask are offline. Cc: Stefan Haberland <[email protected]> Tested-by: Christian Borntraeger <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Fixes: 20e4d81393196 ("blk-mq: simplify queue mapping & schedule with each possisble CPU") Cc: [email protected] Signed-off-by: Ming Lei <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10loop: fix LOOP_GET_STATUS lock imbalanceOmar Sandoval1-15/+18
Commit 2d1d4c1e591f made loop_get_status() drop lo_ctx_mutex before returning, but the loop_get_status_old(), loop_get_status64(), and loop_get_status_compat() wrappers don't call loop_get_status() if the passed argument is NULL. The callers expect that the lock is dropped, so make sure we drop it in that case, too. Reported-by: [email protected] Fixes: 2d1d4c1e591f ("loop: don't call into filesystem while holding lo_ctl_mutex") Signed-off-by: Omar Sandoval <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10block/loop: fix deadlock after loop_set_statusTetsuo Handa1-4/+8
syzbot is reporting deadlocks at __blkdev_get() [1]. ---------------------------------------- [ 92.493919] systemd-udevd D12696 525 1 0x00000000 [ 92.495891] Call Trace: [ 92.501560] schedule+0x23/0x80 [ 92.502923] schedule_preempt_disabled+0x5/0x10 [ 92.504645] __mutex_lock+0x416/0x9e0 [ 92.510760] __blkdev_get+0x73/0x4f0 [ 92.512220] blkdev_get+0x12e/0x390 [ 92.518151] do_dentry_open+0x1c3/0x2f0 [ 92.519815] path_openat+0x5d9/0xdc0 [ 92.521437] do_filp_open+0x7d/0xf0 [ 92.527365] do_sys_open+0x1b8/0x250 [ 92.528831] do_syscall_64+0x6e/0x270 [ 92.530341] entry_SYSCALL_64_after_hwframe+0x42/0xb7 [ 92.931922] 1 lock held by systemd-udevd/525: [ 92.933642] #0: 00000000a2849e25 (&bdev->bd_mutex){+.+.}, at: __blkdev_get+0x73/0x4f0 ---------------------------------------- The reason of deadlock turned out that wait_event_interruptible() in blk_queue_enter() got stuck with bdev->bd_mutex held at __blkdev_put() due to q->mq_freeze_depth == 1. ---------------------------------------- [ 92.787172] a.out S12584 634 633 0x80000002 [ 92.789120] Call Trace: [ 92.796693] schedule+0x23/0x80 [ 92.797994] blk_queue_enter+0x3cb/0x540 [ 92.803272] generic_make_request+0xf0/0x3d0 [ 92.807970] submit_bio+0x67/0x130 [ 92.810928] submit_bh_wbc+0x15e/0x190 [ 92.812461] __block_write_full_page+0x218/0x460 [ 92.815792] __writepage+0x11/0x50 [ 92.817209] write_cache_pages+0x1ae/0x3d0 [ 92.825585] generic_writepages+0x5a/0x90 [ 92.831865] do_writepages+0x43/0xd0 [ 92.836972] __filemap_fdatawrite_range+0xc1/0x100 [ 92.838788] filemap_write_and_wait+0x24/0x70 [ 92.840491] __blkdev_put+0x69/0x1e0 [ 92.841949] blkdev_close+0x16/0x20 [ 92.843418] __fput+0xda/0x1f0 [ 92.844740] task_work_run+0x87/0xb0 [ 92.846215] do_exit+0x2f5/0xba0 [ 92.850528] do_group_exit+0x34/0xb0 [ 92.852018] SyS_exit_group+0xb/0x10 [ 92.853449] do_syscall_64+0x6e/0x270 [ 92.854944] entry_SYSCALL_64_after_hwframe+0x42/0xb7 [ 92.943530] 1 lock held by a.out/634: [ 92.945105] #0: 00000000a2849e25 (&bdev->bd_mutex){+.+.}, at: __blkdev_put+0x3c/0x1e0 ---------------------------------------- The reason of q->mq_freeze_depth == 1 turned out that loop_set_status() forgot to call blk_mq_unfreeze_queue() at error paths for info->lo_encrypt_type != NULL case. ---------------------------------------- [ 37.509497] CPU: 2 PID: 634 Comm: a.out Tainted: G W 4.16.0+ #457 [ 37.513608] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 05/19/2017 [ 37.518832] RIP: 0010:blk_freeze_queue_start+0x17/0x40 [ 37.521778] RSP: 0018:ffffb0c2013e7c60 EFLAGS: 00010246 [ 37.524078] RAX: 0000000000000000 RBX: ffff8b07b1519798 RCX: 0000000000000000 [ 37.527015] RDX: 0000000000000002 RSI: ffffb0c2013e7cc0 RDI: ffff8b07b1519798 [ 37.529934] RBP: ffffb0c2013e7cc0 R08: 0000000000000008 R09: 47a189966239b898 [ 37.532684] R10: dad78b99b278552f R11: 9332dca72259d5ef R12: ffff8b07acd73678 [ 37.535452] R13: 0000000000004c04 R14: 0000000000000000 R15: ffff8b07b841e940 [ 37.538186] FS: 00007fede33b9740(0000) GS:ffff8b07b8e80000(0000) knlGS:0000000000000000 [ 37.541168] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 37.543590] CR2: 00000000206fdf18 CR3: 0000000130b30006 CR4: 00000000000606e0 [ 37.546410] Call Trace: [ 37.547902] blk_freeze_queue+0x9/0x30 [ 37.549968] loop_set_status+0x67/0x3c0 [loop] [ 37.549975] loop_set_status64+0x3b/0x70 [loop] [ 37.549986] lo_ioctl+0x223/0x810 [loop] [ 37.549995] blkdev_ioctl+0x572/0x980 [ 37.550003] block_ioctl+0x34/0x40 [ 37.550006] do_vfs_ioctl+0xa7/0x6d0 [ 37.550017] ksys_ioctl+0x6b/0x80 [ 37.573076] SyS_ioctl+0x5/0x10 [ 37.574831] do_syscall_64+0x6e/0x270 [ 37.576769] entry_SYSCALL_64_after_hwframe+0x42/0xb7 ---------------------------------------- [1] https://syzkaller.appspot.com/bug?id=cd662bc3f6022c0979d01a262c318fab2ee9b56f Signed-off-by: Tetsuo Handa <[email protected]> Reported-by: syzbot <bot+48594378e9851eab70bcd6f99327c7db58c5a28a@syzkaller.appspotmail.com> Fixes: ecdd09597a572513 ("block/loop: fix race between I/O and set_status") Cc: Ming Lei <[email protected]> Cc: Dmitry Vyukov <[email protected]> Cc: stable <[email protected]> Cc: Jens Axboe <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10blk-mq: order getting budget and driver tagMing Lei1-11/+10
This patch orders getting budget and driver tag by making sure to acquire driver tag after budget is got, this way can help to avoid the following race: 1) before dispatch request from scheduler queue, get one budget first, then dequeue a request, call it request A. 2) in another IO path for dispatching request B which is from hctx->dispatch, driver tag is got, then try to get budget in blk_mq_dispatch_rq_list(), unfortunately the budget is held by request A. 3) meantime blk_mq_dispatch_rq_list() is called for dispatching request A, and try to get driver tag first, unfortunately no driver tag is available because the driver tag is held by request B 4) both two IO pathes can't move on, and IO stall is caused. This issue can be observed when running dbench on USB storage. This patch fixes this issue by always getting budget before getting driver tag. Cc: [email protected] Fixes: de1482974080ec9e ("blk-mq: introduce .get_budget and .put_budget in blk_mq_ops") Cc: Christoph Hellwig <[email protected]> Cc: Bart Van Assche <[email protected]> Cc: Omar Sandoval <[email protected]> Signed-off-by: Ming Lei <[email protected]> Signed-off-by: Jens Axboe <[email protected]>
2018-04-10io: change writeX_relaxed() to remove barriersSinan Kaya1-4/+20
Now that we hardened writeX() API in asm-generic version, writeX_relaxed() API is violating the rules when writeX_relaxed() == writeX() in the default implementation. The relaxed API shouldn't have any barriers in it and it doesn't provide any ordering with respect to the memory transactions. The only requirement is for writes to be ordered with respect to each other. This is achieved by the volatile in the __raw_writeX() API. Open code the relaxed API and remove any barriers in it. Signed-off-by: Sinan Kaya <[email protected]> Signed-off-by: Arnd Bergmann <[email protected]>
2018-04-10io: change readX_relaxed() to remove barriersSinan Kaya1-4/+20
Now that we hardened readX() API in asm-generic version, readX_relaxed() API is violating the rules when readX_relaxed() == readX() in the default implementation. The relaxed API shouldn't have any barriers in it and it doesn't provide any ordering with respect to the memory transactions. The only requirement is for reads to be ordered with respect to each other. This is achieved by the volatile in the __raw_readX() API. Open code the relaxed API and remove any barriers in it. Signed-off-by: Sinan Kaya <[email protected]> Signed-off-by: Arnd Bergmann <[email protected]>
2018-04-10x86/cpu: Prevent cpuinfo_x86::x86_phys_bits adjustment corruptionKirill A. Shutemov1-12/+20
Some features (Intel MKTME, AMD SME) reduce the number of effectively available physical address bits. cpuinfo_x86::x86_phys_bits is adjusted accordingly during the early cpu feature detection. Though if get_cpu_cap() is called later again then this adjustement is overwritten. That happens in setup_pku(), which is called after detect_tme(). To address this, extract the address sizes enumeration into a separate function, which is only called only from early_identify_cpu() and from generic_identify(). This makes get_cpu_cap() safe to be called later during boot proccess without overwriting cpuinfo_x86::x86_phys_bits. [ tglx: Massaged changelog ] Fixes: cb06d8e3d020 ("x86/tme: Detect if TME and MKTME is activated by BIOS") Reported-by: Kai Huang <[email protected]> Signed-off-by: Kirill A. Shutemov <[email protected]> Signed-off-by: Thomas Gleixner <[email protected]> Cc: Tom Lendacky <[email protected]> Cc: Dave Hansen <[email protected]> Cc: [email protected] Cc: "H. Peter Anvin" <[email protected]> Link: https://lkml.kernel.org/r/[email protected]
2018-04-11powerpc/powernv: define a standard delay for OPAL_BUSY type retry loopsNicholas Piggin1-0/+3
This is the start of an effort to tidy up and standardise all the delays. Existing loops have a range of delay/sleep periods from 1ms to 20ms, and some have no delay. They all loop forever except rtc, which times out after 10 retries, and that uses 10ms delays. So use 10ms as our standard delay. The OPAL maintainer agrees 10ms is a reasonable starting point. The idea is to use the same recipe everywhere, once this is proven to work then it will be documented as an OPAL API standard. Then both firmware and OS can agree, and if a particular call needs something else, then that can be documented with reasoning. This is not the end-all of this effort, it's just a relatively easy change that fixes some existing high latency delays. There should be provision for standardising timeouts and/or interruptible loops where possible, so non-fatal firmware errors don't cause hangs. Signed-off-by: Nicholas Piggin <[email protected]> Signed-off-by: Michael Ellerman <[email protected]>
2018-04-10c6x: pass endianness info to sparseLuc Van Oostenryck1-0/+1
c6x depends on the macro '_BIG_ENDIAN' being defined or not to correctly select or define endian-specific macros, structures or pieces of code. This macro is predefined by the compiler but sparse knows nothing about it and thus may pre-process files differently from what gcc would. Fix this by passing '-D_BIG_ENDIAN' when compiling a big-endian kernel, like GCC would have done. To: Mark Salter <[email protected]> To: Aurelien Jacquiot <[email protected]> CC: [email protected] Signed-off-by: Luc Van Oostenryck <[email protected]> Signed-off-by: Mark Salter <[email protected]>
2018-04-10c6x: fix platforms/plldata.c get_coreid build errorRandy Dunlap1-0/+1
Fix build error reported by the 0day bot by including the header file for that macro. Fixes this build error: (should fix; not tested) arch/c6x/platforms/plldata.c: In function 'c6472_setup_clocks': arch/c6x/platforms/plldata.c:279:33: error: implicit declaration of function 'get_coreid'; did you mean 'get_order'? [-Werror=implicit-function-declaration] c6x_core_clk.parent = &sysclks[get_coreid() + 1]; Reported-by: Fengguang Wu <[email protected]> Cc: Mark Salter <[email protected]> Cc: Aurelien Jacquiot <[email protected]> Cc: [email protected] Cc: Ingo Molnar <[email protected]> Signed-off-by: Randy Dunlap <[email protected]> Signed-off-by: Mark Salter <[email protected]>
2018-04-10c6x: remove unused KTHREAD_SIZE definitionJérémy Lefaure1-1/+0
KTHREAD_SIZE has never been used since it has been defined for c6x arch. Let's remove this useless definition. Signed-off-by: Jérémy Lefaure <[email protected]> Signed-off-by: Mark Salter <[email protected]>
2018-04-10xen/pvh: Indicate XENFEAT_linux_rsdp_unrestricted to XenBoris Ostrovsky2-1/+26
Pre-4.17 kernels ignored start_info's rsdp_paddr pointer and instead relied on finding RSDP in standard location in BIOS RO memory. This has worked since that's where Xen used to place it. However, with recent Xen change (commit 4a5733771e6f ("libxl: put RSDP for PVH guest near 4GB")) it prefers to keep RSDP at a "non-standard" address. Even though as of commit b17d9d1df3c3 ("x86/xen: Add pvh specific rsdp address retrieval function") Linux is able to find RSDP, for back-compatibility reasons we need to indicate to Xen that we can handle this, an we do so by setting XENFEAT_linux_rsdp_unrestricted flag in ELF notes. (Also take this opportunity and sync features.h header file with Xen) Signed-off-by: Boris Ostrovsky <[email protected]> Reviewed-by: Juergen Gross <[email protected]> Reviewed-by: Wei Liu <[email protected]>
2018-04-10tick-sched: avoid a maybe-uninitialized warningArnd Bergmann1-3/+6
The use of bitfields seems to confuse gcc, leading to a false-positive warning in all compiler versions: kernel/time/tick-sched.c: In function 'tick_nohz_idle_exit': kernel/time/tick-sched.c:538:2: error: 'now' may be used uninitialized in this function [-Werror=maybe-uninitialized] This introduces a temporary variable to track the flags so gcc doesn't have to evaluate twice, eliminating the code path that leads to the warning. Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85301 Fixes: 1cae544d42d2 ("nohz: Gather tick_sched booleans under a common flag field") Signed-off-by: Arnd Bergmann <[email protected]> Signed-off-by: Rafael J. Wysocki <[email protected]>
2018-04-10cpufreq: Drop cpufreq_table_validate_and_show()Viresh Kumar4-27/+7
This isn't used anymore. Remove the helper and update documentation accordingly. Signed-off-by: Viresh Kumar <[email protected]> Signed-off-by: Rafael J. Wysocki <[email protected]>
2018-04-10cpufreq: SCMI: Don't validate the frequency table twiceViresh Kumar1-9/+1
The cpufreq core is already validating the CPU frequency table after calling the ->init() callback of the cpufreq drivers and the drivers don't need to do the same anymore. Though they need to set the policy->freq_table field directly from the ->init() callback now. Stop validating the frequency table from SCMI driver. Signed-off-by: Viresh Kumar <[email protected]> Acked-by: Sudeep Holla <[email protected]> Signed-off-by: Rafael J. Wysocki <[email protected]>
2018-04-10cpufreq: CPPC: Initialize shared perf capabilities of CPUsShunyong Yang1-2/+12
When multiple CPUs are related in one cpufreq policy, the first online CPU will be chosen by default to handle cpufreq operations. Let's take cpu0 and cpu1 as an example. When cpu0 is offline, policy->cpu will be shifted to cpu1. cpu1's perf capabilities should be initialized. Otherwise, perf capabilities are 0s and speed change can not take effect. This patch copies perf capabilities of the first online CPU to other shared CPUs when policy shared type is CPUFREQ_SHARED_TYPE_ANY. Acked-by: Viresh Kumar <[email protected]> Signed-off-by: Shunyong Yang <[email protected]> Signed-off-by: Rafael J. Wysocki <[email protected]>
2018-04-10cpufreq: armada-37xx: Fix clock leakGregory CLEMENT1-0/+2
There was no clk_put() balancing the clk_get(). This commit fixes it. Fixes: 92ce45fb875d (cpufreq: Add DVFS support for Armada 37xx) Cc: 4.16+ <[email protected]> # 4.16+ Reported-by: Thomas Petazzoni <[email protected]> Signed-off-by: Gregory CLEMENT <[email protected]> Acked-by: Viresh Kumar <[email protected]> Signed-off-by: Rafael J. Wysocki <[email protected]>
2018-04-10cpufreq: CPPC: Don't set transition_latencyViresh Kumar1-1/+0
Now that the driver has started to set transition_delay_us directly, there is no need to set transition_latency along with it, as it is not used by the cpufreq core. Signed-off-by: Viresh Kumar <[email protected]> Signed-off-by: Rafael J. Wysocki <[email protected]>
2018-04-10cpufreq: ti-cpufreq: Use builtin_platform_driver()Viresh Kumar1-1/+1
This driver can not be built as a module and there is no need of the platform driver unregister part. Use builtin_platform_driver() instead of module_platform_driver(). Signed-off-by: Viresh Kumar <[email protected]> Signed-off-by: Rafael J. Wysocki <[email protected]>
2018-04-10cpufreq: intel_pstate: Do not include debugfs.hRafael J. Wysocki1-1/+0
The intel_pstate driver doesn't use debugfs any more, so drop linux/debugfs.h from the list of included headers in it. Signed-off-by: Rafael J. Wysocki <[email protected]> Acked-by: Viresh Kumar <[email protected]>
2018-04-09Merge remote-tracking branch ↵Benson Leung4-580/+579
'chrome-platform-stop-being-a-platform-driver-plus-atmel_mxt_ts-for-v4.17' into working-branch-for-4.17
2018-04-09platform/chrome: cros_ec_lpc: Add support for Google devices using custom ↵Salvatore Bellizzi1-0/+12
coreboot firmware This patch adds generic device information to the DMI table of the cros_ec_lpc driver, needed for Chromebooks/boxes using a custom coreboot firmware. The DMI info would not contain "Google_*" as BIOS version string, instead the system vendor string would still be "GOOGLE", so this seems to be a reasonable match for every Chromebook/box running a custom firmware. Signed-off-by: Salvatore Bellizzi <[email protected]> Signed-off-by: Vittorio Gambaletta <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-04-09platform/chrome: cros_ec_lpc: wake up from s2idle on Chrome ECWenkai Du1-0/+4
Chrome platform installed a Chrome EC notify handler which prevents default EC GPE handler getting called. Add pm_system_wakeup to the Chrome EC notify handler so wake up from s2idle can happen. Suggested-by: Rafael J. Wysocki <[email protected]> Signed-off-by: Wenkai Du <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-04-09Input: atmel_mxt_ts - remove platform data supportDmitry Torokhov3-107/+55
Now that there are no users of custom Atmel platform data, and everyone has switched to the generic device properties, we can remove support for the platform data. Acked-by: Nick Dyer <[email protected]> Signed-off-by: Dmitry Torokhov <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-04-09platform/chrome: chromeos_laptop - discard data for unneeded boardsDmitry Torokhov1-212/+264
Mark board data as __intconst/__initdata and make a copy of appropriate entry once we identified the board we are running on. The rest of the data will be discarded once the kernel finished booting (or module finished loading). Signed-off-by: Dmitry Torokhov <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-04-09platform/chrome: chromeos_laptop - use device properties for PixelDmitry Torokhov1-6/+6
Now that Atmel driver uses generic device properties we can use them instead of platform data when setting up touchpad on the original Google Pixel. Signed-off-by: Dmitry Torokhov <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-04-09platform/chrome: chromeos_laptop - rely on I2C to set up interrupt triggerDmitry Torokhov1-7/+14
Instead of passing interrupt flags via platform data to drivers, or hoping that drivers will do the right thing and set it up the way we need, let's set up IRQ resource and attach it to the I2C board info, and let I2C core set it up for us. Signed-off-by: Dmitry Torokhov <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-04-09platform/chrome: chromeos_laptop - use I2C notifier to create devicesDmitry Torokhov1-145/+102
Instead of using platform device and deferrals to handle the case when i2C adapters appear late in the game, and not handling device unbinding all that well, let's switch to using I2C bus notifier to get told when a new I2C adapter appears in the system, and attempt to add appropriate devices at that time. In case when we have 2 Designware adapters in the system (Acer C720), instead of counting and hoping they get enumerate din the right order, let's switch to using their PCI devids (slot/function) that should be stable. Signed-off-by: Dmitry Torokhov <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-04-09platform/chrome: chromeos_laptop - parse DMI IRQ data onceDmitry Torokhov1-56/+64
Instead of trying to parse DMI IRQ data every time we try to instantiate a device, let's do it once, when we identify the device we are working with. This allows us to mark chromeos_laptop_get_irq_from_dmi() as __init and discard it once module is initialized. Signed-off-by: Dmitry Torokhov <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-04-09platform/chrome: chromeos_laptop - rework i2c peripherals initializationDmitry Torokhov1-232/+235
Instead of having separate setup() functions responsible for instantiating i2c client for each peripheral, let's generalize the behavior and use common code for instantiating all i2c peripherals. Signed-off-by: Dmitry Torokhov <[email protected]> Signed-off-by: Benson Leung <[email protected]>
2018-04-10s390/zcrypt: Remove deprecated zcrypt proc interface.Harald Freudenberger1-231/+0
This patch removes the deprecated zcrypt proc interface. It is outdated and deprecated and does not support the latest 3 generations of CEX cards. Signed-off-by: Harald Freudenberger <[email protected]> Signed-off-by: Martin Schwidefsky <[email protected]>
2018-04-10s390/zcrypt: Remove deprecated ioctls.Harald Freudenberger3-125/+2
This patch removes the old status calls which have been marked as deprecated since at least 2 years now. There is no known application or library relying on these ioctls any more. Signed-off-by: Harald Freudenberger <[email protected]> Signed-off-by: Martin Schwidefsky <[email protected]>
2018-04-10s390/zcrypt: Make ap init functions static.Harald Freudenberger2-5/+2
The ap init functions ap_module_init and ap_debug_init are only used within ap_bus.c. Make these functions static and do not declare them in any header file any more. Signed-off-by: Harald Freudenberger <[email protected]> Signed-off-by: Martin Schwidefsky <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.