November 2018 - Linux-stable-mirror

[PATCH] media: rc: cec devices do not have a lirc chardev

by Sean Young

On Mon, Oct 22, 2018 at 11:44:22AM +0200, Torbjorn Jansson wrote: > On 2018-10-22 10:59, Sean Young wrote: > > On Sat, Oct 20, 2018 at 11:12:16PM +0200, Hans Verkuil wrote: > > > Hi Sean, > > > > > > Can you take a look at this, it appears to be an RC issue, see my analysis below. > > > > > > On 10/20/2018 03:26 PM, Torbjorn Jansson wrote: > > > > Hello > > > > > > > > i'm using the pulse8 usb cec adapter to control my tv. > > > > i have a few scripts that poll the power status of my tv and after a while it stops working returning errors when trying to check if tv is on or off. > > > > this i think matches a kernel oops i'm seeing that i suspect is related to this. > > > > > > > > i have sometimes been able to recover from this problem by completely cutting power to my tv and also unplugging the usb cec adapter. > > > > i have a feeling that the tv is at least partly to blame for cec-ctl not working but in any case there shouldn't be a kernel oops. > > > > > > > > > > > > also every now and then i see this in dmesg: > > > > cec cec0: transmit: failed 05 > > > > cec cec0: transmit: failed 06 > > > > but that doesn't appear to do any harm as far as i can tell. > > > > > > > > any idea whats causing the oops? > > > > > > > > the ops: > > > > > > > > BUG: unable to handle kernel NULL pointer dereference at 0000000000000038 > > > > PGD 0 P4D 0 > > > > Oops: 0000 [#1] SMP PTI > > > > CPU: 9 PID: 27687 Comm: kworker/9:2 Tainted: P OE 4.18.12-200.fc28.x86_64 #1 > > > > Hardware name: Supermicro C7X99-OCE-F/C7X99-OCE-F, BIOS 2.1a 06/15/2018 > > > > Workqueue: events pulse8_irq_work_handler [pulse8_cec] > > > > RIP: 0010:ir_lirc_scancode_event+0x3d/0xb0 [rc_core] > > > > > > Huh. ir_lirc_scancode_event() calls spin_lock_irqsave(&dev->lirc_fh_lock, flags); > > > > > > The spinlock dev->lirc_fh_lock is initialized in ir_lirc_register(), which is called > > > from rc_register_device(), except when the protocol is CEC: > > > > > > /* Ensure that the lirc kfifo is setup before we start the thread */ > > > if (dev->allowed_protocols != RC_PROTO_BIT_CEC) { > > > rc = ir_lirc_register(dev); > > > if (rc < 0) > > > goto out_rx; > > > } > > > > > > So it looks like ir_lirc_scancode_event() fails because dev->lirc_fh_lock was never > > > initialized. > > > > > > Could this be fall-out from the lirc changes you did not too long ago? > > > > Yes, this is broken. My bad, sorry. I think this must have been broken since > > v4.16. I can write a patch but I don't have a patch but I'm on the train > > to ELCE in Edinburgh now, with no hardware to test on. > > > > > > Sean > > > > the kernel oops have been happening for a while now. > yesterday when i checked my logs i can see them at least back a couple of > months when i was running 4.17 > > also my scripts to poll status of my tv and turn it on/off works for a while > so it doesn't crash right away. > maybe it only crashes when i send cec command to turn on/off tv and only > polling for status is no problem. > > > i think i have a separate issue too because i had problems even before the > kernel oopses started. > but i suspect this is caused by my tv locking up the cec bus because > unplugging power to tv for a few minutes (i must wait or it will still be > just as broken) and then back used to resolve the cec errors from my > scripts. Would you be able to test the following patch please? Thanks, Sean --- >From 1b8b20b606b30c0e301c80e18af8d77194269bc1 Mon Sep 17 00:00:00 2001 From: Sean Young <sean(a)mess.org> Date: Mon, 22 Oct 2018 10:01:50 +0100 Subject: [PATCH] media: rc: cec devices do not have a lirc chardev MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This fixes an oops in ir_lirc_scancode_event(). BUG: unable to handle kernel NULL pointer dereference at 0000000000000038 PGD 0 P4D 0 Oops: 0000 [#1] SMP PTI CPU: 9 PID: 27687 Comm: kworker/9:2 Tainted: P OE 4.18.12-200.fc28.x86_64 #1 Hardware name: Supermicro C7X99-OCE-F/C7X99-OCE-F, BIOS 2.1a 06/15/2018 Workqueue: events pulse8_irq_work_handler [pulse8_cec] RIP: 0010:ir_lirc_scancode_event+0x3d/0xb0 [rc_core] Code: 8d ae b4 07 00 00 49 81 c6 b8 07 00 00 53 e8 4a df c3 d5 48 89 ef 49 89 45 00 e8 4e 84 41 d6 49 8b 1e 49 89 c4 4c 39 f3 74 58 <8b> 43 38 8b 53 40 89 c1 2b 4b 3c 39 ca 72 41 21 d0 49 8b 7d 00 49 RSP: 0018:ffffaa10e3c07d58 EFLAGS: 00010017 RAX: 0000000000000002 RBX: 0000000000000000 RCX: 0000000000000018 RDX: 0000000000000001 RSI: 00316245397fa93c RDI: ffff966d31c8d7b4 RBP: ffff966d31c8d7b4 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000003 R11: ffffaa10e3c07e28 R12: 0000000000000002 R13: ffffaa10e3c07d88 R14: ffff966d31c8d7b8 R15: 0000000000000073 FS: 0000000000000000(0000) GS:ffff966d3f440000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000038 CR3: 00000009d820a003 CR4: 00000000003606e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: ir_do_keydown+0x75/0x260 [rc_core] rc_keydown+0x54/0xc0 [rc_core] cec_received_msg_ts+0xaa8/0xaf0 [cec] process_one_work+0x1a1/0x350 worker_thread+0x30/0x380 ? pwq_unbound_release_workfn+0xd0/0xd0 kthread+0x112/0x130 ? kthread_create_worker_on_cpu+0x70/0x70 ret_from_fork+0x35/0x40 Modules linked in: rc_tt_1500 dvb_usb_dvbsky dvb_usb_v2 uas usb_storage fuse vhost_net vhost tap xt_CHECKSUM iptable_mangle ip6t_REJECT nf_reject_ipv6 tun 8021q garp mrp xt_nat macvlan xfs devlink ebta si2157 si2168 cx25840 cx23885 kvm altera_ci tda18271 joydev ir_rc6_decoder rc_rc6_mce crct10dif_pclmul crc32_pclmul ghash_clmulni_intel intel_cstate intel_uncore altera_stapl m88ds3103 tveeprom cx2341 mxm_wmi igb crc32c_intel megaraid_sas dca i2c_algo_bit wmi vfio_pci irqbypass vfio_virqfd vfio_iommu_type1 vfio i2c_dev CR2: 0000000000000038 Cc: <stable(a)vger.kernel.org> # v4.16+ Signed-off-by: Sean Young <sean(a)mess.org> --- drivers/media/rc/rc-main.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/media/rc/rc-main.c b/drivers/media/rc/rc-main.c index 552bbe82a160..877978dbd409 100644 --- a/drivers/media/rc/rc-main.c +++ b/drivers/media/rc/rc-main.c @@ -695,7 +695,8 @@ void rc_repeat(struct rc_dev *dev) (dev->last_toggle ? LIRC_SCANCODE_FLAG_TOGGLE : 0) }; - ir_lirc_scancode_event(dev, &sc); + if (dev->allowed_protocols != RC_PROTO_BIT_CEC) + ir_lirc_scancode_event(dev, &sc); spin_lock_irqsave(&dev->keylock, flags); @@ -735,7 +736,8 @@ static void ir_do_keydown(struct rc_dev *dev, enum rc_proto protocol, .keycode = keycode }; - ir_lirc_scancode_event(dev, &sc); + if (dev->allowed_protocols != RC_PROTO_BIT_CEC) + ir_lirc_scancode_event(dev, &sc); if (new_event && dev->keypressed) ir_do_keyup(dev, false); -- 2.17.2

6 years, 1 month

3
5
0 0

v4.19.2 build: 0 failures 3 warnings (v4.19.2)

by Build bot for Mark Brown

Tree/Branch: v4.19.2 Git describe: v4.19.2 Commit: 7950eb316a Linux 4.19.2 Build Time: 124 min 27 sec Passed: 11 / 11 (100.00 %) Failed: 0 / 11 ( 0.00 %) Errors: 0 Warnings: 3 Section Mismatches: 0 ------------------------------------------------------------------------------- defconfigs with issues (other than build errors): 1 warnings 0 mismatches : arm64-allmodconfig 1 warnings 0 mismatches : arm-allmodconfig 1 warnings 0 mismatches : arm-multi_v5_defconfig ------------------------------------------------------------------------------- Warnings Summary: 3 1 ../drivers/staging/erofs/unzip_vle.c:186:29: warning: array subscript is above array bounds [-Warray-bounds] 1 ../drivers/isdn/hardware/eicon/message.c:5985:1: warning: the frame size of 2064 bytes is larger than 2048 bytes [-Wframe-larger-than=] 1 ../drivers/i2c/busses/i2c-aspeed.c:567:1: warning: label 'out' defined but not used [-Wunused-label] =============================================================================== Detailed per-defconfig build reports below: ------------------------------------------------------------------------------- arm64-allmodconfig : PASS, 0 errors, 1 warnings, 0 section mismatches Warnings: ../drivers/isdn/hardware/eicon/message.c:5985:1: warning: the frame size of 2064 bytes is larger than 2048 bytes [-Wframe-larger-than=] ------------------------------------------------------------------------------- arm-allmodconfig : PASS, 0 errors, 1 warnings, 0 section mismatches Warnings: ../drivers/staging/erofs/unzip_vle.c:186:29: warning: array subscript is above array bounds [-Warray-bounds] ------------------------------------------------------------------------------- arm-multi_v5_defconfig : PASS, 0 errors, 1 warnings, 0 section mismatches Warnings: ../drivers/i2c/busses/i2c-aspeed.c:567:1: warning: label 'out' defined but not used [-Wunused-label] ------------------------------------------------------------------------------- Passed with no errors, warnings or mismatches: arm64-allnoconfig arm-multi_v7_defconfig x86_64-defconfig arm-allnoconfig x86_64-allnoconfig arm-multi_v4t_defconfig x86_64-allmodconfig arm64-defconfig

6 years, 1 month

1
0
0 0

[PATCH 1/2] drm/i915: Mark up GTT sizes as u64

by Chris Wilson

Since we use a 64b virtual GTT irrespective of the system, we want to ensure that the GTT computations remains 64b even on 32b systems, including treatment of huge virtual pages. No code generation changes on 64b: Reported-by: Sergii Romantsov <sergii.romantsov(a)globallogic.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108282 Signed-off-by: Chris Wilson <chris(a)chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen(a)linux.intel.com> Cc: stable(a)vger.kernel.org --- drivers/gpu/drm/i915/i915_gem_gtt.c | 2 +- drivers/gpu/drm/i915/i915_gem_gtt.h | 6 +++--- drivers/gpu/drm/i915/selftests/huge_pages.c | 2 +- drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 6 +++--- 4 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 98d9a1eb1ed2..19b2d991b5d8 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -1708,7 +1708,7 @@ static void gen6_dump_ppgtt(struct i915_hw_ppgtt *base, struct seq_file *m) if (i == 4) continue; - seq_printf(m, "\t\t(%03d, %04d) %08lx: ", + seq_printf(m, "\t\t(%03d, %04d) %08llx: ", pde, pte, (pde * GEN6_PTES + pte) * I915_GTT_PAGE_SIZE); for (i = 0; i < 4; i++) { diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h index 7e2af5f4f39b..aa8307043036 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.h +++ b/drivers/gpu/drm/i915/i915_gem_gtt.h @@ -42,9 +42,9 @@ #include "i915_selftest.h" #include "i915_timeline.h" -#define I915_GTT_PAGE_SIZE_4K BIT(12) -#define I915_GTT_PAGE_SIZE_64K BIT(16) -#define I915_GTT_PAGE_SIZE_2M BIT(21) +#define I915_GTT_PAGE_SIZE_4K BIT_ULL(12) +#define I915_GTT_PAGE_SIZE_64K BIT_ULL(16) +#define I915_GTT_PAGE_SIZE_2M BIT_ULL(21) #define I915_GTT_PAGE_SIZE I915_GTT_PAGE_SIZE_4K #define I915_GTT_MAX_PAGE_SIZE I915_GTT_PAGE_SIZE_2M diff --git a/drivers/gpu/drm/i915/selftests/huge_pages.c b/drivers/gpu/drm/i915/selftests/huge_pages.c index 09ea65a29d98..256001b00e32 100644 --- a/drivers/gpu/drm/i915/selftests/huge_pages.c +++ b/drivers/gpu/drm/i915/selftests/huge_pages.c @@ -551,7 +551,7 @@ static int igt_mock_ppgtt_misaligned_dma(void *arg) err = igt_check_page_sizes(vma); if (vma->page_sizes.gtt != I915_GTT_PAGE_SIZE_4K) { - pr_err("page_sizes.gtt=%u, expected %lu\n", + pr_err("page_sizes.gtt=%u, expected %llu\n", vma->page_sizes.gtt, I915_GTT_PAGE_SIZE_4K); err = -EINVAL; } diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index 17b5aaaa7a50..69fe86b30fbb 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -1337,7 +1337,7 @@ static int igt_gtt_reserve(void *arg) GEM_BUG_ON(!drm_mm_node_allocated(&vma->node)); if (vma->node.start != total || vma->node.size != 2*I915_GTT_PAGE_SIZE) { - pr_err("i915_gem_gtt_reserve (pass 1) placement failed, found (%llx + %llx), expected (%llx + %lx)\n", + pr_err("i915_gem_gtt_reserve (pass 1) placement failed, found (%llx + %llx), expected (%llx + %llx)\n", vma->node.start, vma->node.size, total, 2*I915_GTT_PAGE_SIZE); err = -EINVAL; @@ -1386,7 +1386,7 @@ static int igt_gtt_reserve(void *arg) GEM_BUG_ON(!drm_mm_node_allocated(&vma->node)); if (vma->node.start != total || vma->node.size != 2*I915_GTT_PAGE_SIZE) { - pr_err("i915_gem_gtt_reserve (pass 2) placement failed, found (%llx + %llx), expected (%llx + %lx)\n", + pr_err("i915_gem_gtt_reserve (pass 2) placement failed, found (%llx + %llx), expected (%llx + %llx)\n", vma->node.start, vma->node.size, total, 2*I915_GTT_PAGE_SIZE); err = -EINVAL; @@ -1430,7 +1430,7 @@ static int igt_gtt_reserve(void *arg) GEM_BUG_ON(!drm_mm_node_allocated(&vma->node)); if (vma->node.start != offset || vma->node.size != 2*I915_GTT_PAGE_SIZE) { - pr_err("i915_gem_gtt_reserve (pass 3) placement failed, found (%llx + %llx), expected (%llx + %lx)\n", + pr_err("i915_gem_gtt_reserve (pass 3) placement failed, found (%llx + %llx), expected (%llx + %llx)\n", vma->node.start, vma->node.size, offset, 2*I915_GTT_PAGE_SIZE); err = -EINVAL; -- 2.19.1

6 years, 1 month

3
4
0 0

[PATCH] usb: dwc3: gadget: fix ISOC TRB type on unaligned transfers

by Felipe Balbi

When chaining ISOC TRBs together, only the first ISOC TRB should be of type ISOC_FIRST, all others should be of type ISOC. This patch fixes that. Fixes: c6267a51639b ("usb: dwc3: gadget: align transfers to wMaxPacketSize") Cc: <stable(a)vger.kernel.org> # v4.11+ Signed-off-by: Felipe Balbi <felipe.balbi(a)linux.intel.com> --- drivers/usb/dwc3/gadget.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index 06e22afdf3d1..9faad896b3a1 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -1081,7 +1081,7 @@ static void dwc3_prepare_one_trb_sg(struct dwc3_ep *dep, /* Now prepare one extra TRB to align transfer size */ trb = &dep->trb_pool[dep->trb_enqueue]; __dwc3_prepare_one_trb(dep, trb, dwc->bounce_addr, - maxp - rem, false, 0, + maxp - rem, false, 1, req->request.stream_id, req->request.short_not_ok, req->request.no_interrupt); @@ -1125,7 +1125,7 @@ static void dwc3_prepare_one_trb_linear(struct dwc3_ep *dep, /* Now prepare one extra TRB to align transfer size */ trb = &dep->trb_pool[dep->trb_enqueue]; __dwc3_prepare_one_trb(dep, trb, dwc->bounce_addr, maxp - rem, - false, 0, req->request.stream_id, + false, 1, req->request.stream_id, req->request.short_not_ok, req->request.no_interrupt); } else if (req->request.zero && req->request.length && @@ -1141,7 +1141,7 @@ static void dwc3_prepare_one_trb_linear(struct dwc3_ep *dep, /* Now prepare one extra TRB to handle ZLP */ trb = &dep->trb_pool[dep->trb_enqueue]; __dwc3_prepare_one_trb(dep, trb, dwc->bounce_addr, 0, - false, 0, req->request.stream_id, + false, 1, req->request.stream_id, req->request.short_not_ok, req->request.no_interrupt); } else { -- 2.19.1

6 years, 1 month

1
0
0 0

[PATCH] rtc: m41t80: Correct alarm month range with RTC reads

by Maciej W. Rozycki

Add the missing adjustment of the month range on alarm reads from the RTC, correcting an issue coming from commit 9c6dfed92c3e ("rtc: m41t80: add alarm functionality"). The range is 1-12 for hardware and 0-11 for `struct rtc_time', and is already correctly handled on alarm writes to the RTC. It was correct up until commit 48e9766726eb ("drivers/rtc/rtc-m41t80.c: remove disabled alarm functionality") too, which removed the previous implementation of alarm support. Signed-off-by: Maciej W. Rozycki <macro(a)linux-mips.org> Fixes: 9c6dfed92c3e ("rtc: m41t80: add alarm functionality") References: 48e9766726eb ("drivers/rtc/rtc-m41t80.c: remove disabled alarm functionality") Cc: stable(a)vger.kernel.org # 4.7+ --- Hi, I have no means to verify this change at run time as my M41T81 device, which is the reason for me to use this driver, regrettably hasn't got its IRQ line routed, due to historical reasons I gathered, as it replaced a different RTC device, that didn't have an IRQ output at all, used with earlier revisions of the Broadcom SiByte SWARM board I have been using this code with. However it should be obvious by code inspection. Please apply and backport as appropriate. Maciej --- drivers/rtc/rtc-m41t80.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) linux-rtc-m41t80-read-alarm-month.diff Index: linux-20181008-swarm64-eb/drivers/rtc/rtc-m41t80.c =================================================================== --- linux-20181008-swarm64-eb.orig/drivers/rtc/rtc-m41t80.c +++ linux-20181008-swarm64-eb/drivers/rtc/rtc-m41t80.c @@ -393,7 +393,7 @@ static int m41t80_read_alarm(struct devi alrm->time.tm_min = bcd2bin(alarmvals[3] & 0x7f); alrm->time.tm_hour = bcd2bin(alarmvals[2] & 0x3f); alrm->time.tm_mday = bcd2bin(alarmvals[1] & 0x3f); - alrm->time.tm_mon = bcd2bin(alarmvals[0] & 0x3f); + alrm->time.tm_mon = bcd2bin(alarmvals[0] & 0x3f) - 1; alrm->enabled = !!(alarmvals[0] & M41T80_ALMON_AFE); alrm->pending = (flags & M41T80_FLAGS_AF) && alrm->enabled;

6 years, 1 month

2
1
0 0

[PATCH 01/14] usb: dwc3: core: Clean up ULPI device

by Felipe Balbi

From: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> If dwc3_core_init_mode() fails with deferred probe, next probe fails on sysfs with sysfs: cannot create duplicate filename '/devices/pci0000:00/0000:00:11.0/dwc3.0.auto/dwc3.0.auto.ulpi' To avoid this failure, clean up ULPI device. Cc: <stable(a)vger.kernel.org> Signed-off-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Signed-off-by: Felipe Balbi <felipe.balbi(a)linux.intel.com> --- drivers/usb/dwc3/core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c index 75cff57b2629..8f6d9c6f016e 100644 --- a/drivers/usb/dwc3/core.c +++ b/drivers/usb/dwc3/core.c @@ -1505,6 +1505,7 @@ static int dwc3_probe(struct platform_device *pdev) err5: dwc3_event_buffers_cleanup(dwc); + dwc3_ulpi_exit(dwc); err4: dwc3_free_scratch_buffers(dwc); -- 2.19.1

6 years, 1 month

1
2
0 0

[RFC PATCH] SCSI: fix queue cleanup race before queue is initialized done

by Ming Lei

c2856ae2f315d ("blk-mq: quiesce queue before freeing queue") has already fixed this race, however the implied synchronize_rcu() in blk_mq_quiesce_queue() can slow down LUN probe a lot, so caused performance regression. Then 1311326cf4755c7 ("blk-mq: avoid to synchronize rcu inside blk_cleanup_queue()") tried to only quiesce queue for avoiding unnecessary synchronize_rcu() in case that queue isn't initialized done. However, turns out we still need to quiesce the queue in case that queue isn't initialized done. Because when one SCSI command is completed, the user is waken up immediately, then the scsi device can be removed, meantime the run queue in scsi_end_request() can be still in-progress, so kernel panic is triggered. In Red Hat QE lab, there are several reports about this kind of kernel panic triggered during kernel booting. Fixes: 1311326cf4755c7 ("blk-mq: avoid to synchronize rcu inside blk_cleanup_queue()") Cc: Andrew Jones <drjones(a)redhat.com> Cc: Bart Van Assche <bart.vanassche(a)wdc.com> Cc: linux-scsi(a)vger.kernel.org Cc: Martin K. Petersen <martin.petersen(a)oracle.com> Cc: Christoph Hellwig <hch(a)lst.de> Cc: James E.J. Bottomley <jejb(a)linux.vnet.ibm.com> Cc: stable <stable(a)vger.kernel.org> Signed-off-by: Ming Lei <ming.lei(a)redhat.com> --- block/blk-core.c | 6 +++--- drivers/scsi/scsi_lib.c | 36 ++++++++++++++++++++++++++++++------ 2 files changed, 33 insertions(+), 9 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index ce12515f9b9b..cf7742a677c4 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -798,9 +798,9 @@ void blk_cleanup_queue(struct request_queue *q) * dispatch may still be in-progress since we dispatch requests * from more than one contexts. * - * No need to quiesce queue if it isn't initialized yet since - * blk_freeze_queue() should be enough for cases of passthrough - * request. + * We rely on driver to deal with the race in case that queue + * initialization isn't done. + * */ if (q->mq_ops && blk_queue_init_done(q)) blk_mq_quiesce_queue(q); diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index c7fccbb8f554..7ec7a8a2d000 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -697,13 +697,37 @@ static bool scsi_end_request(struct request *req, blk_status_t error, */ scsi_mq_uninit_cmd(cmd); - __blk_mq_end_request(req, error); + /* + * When block queue initialization isn't done, the request + * queue won't be quiesced in blk_cleanup_queue() for avoiding + * slowing down LUN probe, so queue still may be run even though + * its resource is cleaned up, this way can cause kernel panic. + * + * Workaround this issue by freeing request after running the + * queue when queue initialization isn't done, so the queue's + * usage counter can be held during running queue. + * + * This way is safe because sdev->device_busy has been decreased + * already, and scsi_queue_rq() may guarantee the forward-progress. + * + */ + if (blk_queue_init_done(q)) { + __blk_mq_end_request(req, error); + + if (scsi_target(sdev)->single_lun || + !list_empty(&sdev->host->starved_list)) + kblockd_schedule_work(&sdev->requeue_work); + else + blk_mq_run_hw_queues(q, true); + } else { - if (scsi_target(sdev)->single_lun || - !list_empty(&sdev->host->starved_list)) - kblockd_schedule_work(&sdev->requeue_work); - else - blk_mq_run_hw_queues(q, true); + if (scsi_target(sdev)->single_lun || + !list_empty(&sdev->host->starved_list)) + kblockd_schedule_work(&sdev->requeue_work); + else + blk_mq_run_hw_queues(q, true); + __blk_mq_end_request(req, error); + } } else { unsigned long flags; -- 2.9.5

6 years, 1 month

2
2
0 0

PASS: Test report for kernel 4.19.2.skt (linux-stable-rc)

by CKI

Hello, We ran automated tests on a recent commit from this kernel tree: Kernel repo: git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git Commit: 7950eb316adf Linux 4.19.2 The results of these automated tests are provided below. Overall result: PASSED Patch merge: OK Compile: OK Hardware test: OK Please reply to this email if you have any questions about the tests that we ran or if you have any suggestions on how to make future tests more effective. ,-. ,-. ( C ) ( I ) Continuous `-',-.`-' Kernel ( K ) Integration `-' ______________________________________________________________________________ Compile testing --------------- We compiled the kernel for 4 architectures: x86_64: make options: make INSTALL_MOD_STRIP=1 -j56 targz-pkg configuration: https://artifacts.cki-project.org/builds/x86_64/7950eb316adf792283cac5743df… arm64: make options: make INSTALL_MOD_STRIP=1 -j56 targz-pkg configuration: https://artifacts.cki-project.org/builds/aarch64/7950eb316adf792283cac5743d… powerpc: make options: make INSTALL_MOD_STRIP=1 -j56 targz-pkg configuration: https://artifacts.cki-project.org/builds/ppc64le/7950eb316adf792283cac5743d… powerpc: make options: make INSTALL_MOD_STRIP=1 -j56 targz-pkg configuration: https://artifacts.cki-project.org/builds/ppc64/7950eb316adf792283cac5743dfe…

6 years, 1 month

1
0
0 0

stable/linux-4.14.y boot: 39 boots: 0 failed, 39 passed (v4.14.81)

by kernelci.org bot

stable/linux-4.14.y boot: 39 boots: 0 failed, 39 passed (v4.14.81) Full Boot Summary: https://kernelci.org/boot/all/job/stable/branch/linux-4.14.y/kernel/v4.14.8… Full Build Summary: https://kernelci.org/build/stable/branch/linux-4.14.y/kernel/v4.14.81/ Tree: stable Branch: linux-4.14.y Git Describe: v4.14.81 Git Commit: 2e390c487815669fb9bb35d7ea11883cc10a9b50 Git URL: http://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git Tested: 22 unique boards, 11 SoC families, 9 builds out of 180 --- For more info write to <info(a)kernelci.org>

6 years, 1 month

1
0
0 0

stable/linux-4.18.y boot: 25 boots: 0 failed, 24 passed with 1 untried/unknown (v4.18.19)

by kernelci.org bot

stable/linux-4.18.y boot: 25 boots: 0 failed, 24 passed with 1 untried/unknown (v4.18.19) Full Boot Summary: https://kernelci.org/boot/all/job/stable/branch/linux-4.18.y/kernel/v4.18.1… Full Build Summary: https://kernelci.org/build/stable/branch/linux-4.18.y/kernel/v4.18.19/ Tree: stable Branch: linux-4.18.y Git Describe: v4.18.19 Git Commit: 6559b2338d96ec330887871d47e1ced520bc4f03 Git URL: http://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git Tested: 19 unique boards, 10 SoC families, 8 builds out of 180 --- For more info write to <info(a)kernelci.org>

6 years, 1 month

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror November 2018