Re: linux-next: Tree for May 19 (arm:vexpress qemu failure due to FB patch)

[Guenter Roeck <linux@roeck-us.net>]

On 05/20/2015 12:12 AM, Greg Kroah-Hartman wrote:
> On Tue, May 19, 2015 at 11:02:36PM -0700, Guenter Roeck wrote:
>> On Tue, May 19, 2015 at 05:30:01PM +1000, Stephen Rothwell wrote:
>>> Hi all,
>>>
>>> Changes since 20150518:
>>>
>>> The akpm-current tree gained a conflict against Linus' tree.
>>>
>>> Non-merge commits (relative to Linus' tree): 4713
>>>   4246 files changed, 207445 insertions(+), 103657 deletions(-)
>>>
>>> ----------------------------------------------------------------------------
>>>
>> I get an arm qemu runtime failure with this kernel when testing
>> a vexpress configuration. Kernel log output is
>>
>> Freeing unused kernel memory: 264K (80618000 - 8065a000)
>> random: nonblocking pool is initialized
>> ------------[ cut here ]------------
>> WARNING: CPU: 0 PID: 661 at kernel/time/timer.c:1098 del_timer_sync+0x4c/0x54()
>> ** 20 printk messages dropped ** [<80059184>] (handle_irq_event_percpu) from [<8005928c>] (handle_irq_event+0x44/0x64)
>> ** 127 printk messages dropped ** [<80059184>] (handle_irq_event_percpu) from [<8005928c>] (handle_irq_event+0x44/0x64)
>> ** 128 printk messages dropped ** [<8005928c>] (handle_irq_event) from [<8005becc>] (handle_fasteoi_irq+0xa8/0x180)
>> ** 127 printk messages dropped ** [<8005928c>] (handle_irq_event) from [<8005becc>] (handle_fasteoi_irq+0xa8/0x180)
>> ** 127 printk messages dropped ** [<8005928c>] (handle_irq_event) from [<8005becc>] (handle_fasteoi_irq+0xa8/0x180)
>>
>> and many more of those until the qemu session is aborted. The complete log
>> is at
>> http://server.roeck-us.net:8010/builders/qemu-arm-next/builds/165/steps/qemubuildcommand/logs/stdio
>>
>> In some of the bisect builds I also noticed lots of
>>
>> ** 167 printk messages dropped ** [<80055c00>] (call_console_drivers.constprop.25) from [<80057144>] (console_unlock+0x4c8/0x51c)
>>
>> ---
>> Bisect log:
>>
>> # bad: [47cb9006f5fc8568ff5f9359d8b05e8f5fc91f12] Add linux-next specific files for 20150519
>> # good: [e26081808edadfd257c6c9d81014e3b25e9a6118] Linux 4.1-rc4
>> git bisect start 'HEAD' 'v4.1-rc4'
>> # good: [6835509ff89e1da06d5c017c962aa00f06fd554a] Merge remote-tracking branch 'drm-panel/drm/panel/for-next'
>> git bisect good 6835509ff89e1da06d5c017c962aa00f06fd554a
>> # good: [689fbd5603000238443bf1d91277c36962374b27] Merge remote-tracking branch 'rcu/rcu/next'
>> git bisect good 689fbd5603000238443bf1d91277c36962374b27
>> # bad: [dc5f097e91455245a9ddbff7e9cbb475fd89745c] Merge remote-tracking branch 'staging/staging-next'
>> git bisect bad dc5f097e91455245a9ddbff7e9cbb475fd89745c
>> # good: [a4f18b1c40875d199a1f7a77f08642d38c4e6b21] staging: comedi: ni_stc.h: tidy up RTSI_Trig_Direction_Register and bits
>> git bisect good a4f18b1c40875d199a1f7a77f08642d38c4e6b21
>> # bad: [99bd60ce3d5dca9a87c615805947001b5b864102] Merge remote-tracking branch 'tty/tty-next'
>> git bisect bad 99bd60ce3d5dca9a87c615805947001b5b864102
>> # bad: [77232f791e0605521a978cbe7f79a93df24df374] vt: Don't check KD_GRAPHICS when binding/unbinding
>> git bisect bad 77232f791e0605521a978cbe7f79a93df24df374
>> # good: [b8dccc1b2e178c4d9702c30fadbadff0c093e3d7] tty: hvc_console.c: move assignment out of if () block
>> git bisect good b8dccc1b2e178c4d9702c30fadbadff0c093e3d7
>> # good: [31ada047681834d0d6f04adfb305ff26b52d2823] serial: imx: Constify platform_device_id
>> git bisect good 31ada047681834d0d6f04adfb305ff26b52d2823
>> # good: [853a699739fede12a5465be685f0f366276cb507] serial: tegra: handle race condition on uart rx side
>> git bisect good 853a699739fede12a5465be685f0f366276cb507
>> # good: [bd63364caa8df38bad2b25b11b2a1b849475cce5] vt: add cursor blink interval escape sequence
>> git bisect good bd63364caa8df38bad2b25b11b2a1b849475cce5
>> # bad: [03ac6b3499c8a0b1fbb5c33ad6234a9e860f2614] drivers/tty/nozomi.c: rename CONFIG_MAGIC
>> git bisect bad 03ac6b3499c8a0b1fbb5c33ad6234a9e860f2614
>> # bad: [360ee94b47a80b0e986c9e890f1862d29edf2c45] TTY: msm_smd_tty: Remove unused driver
>> git bisect bad 360ee94b47a80b0e986c9e890f1862d29edf2c45
>> # bad: [27a4c827c34ac4256a190cc9d24607f953c1c459] fbcon: use the cursor blink interval provided by vt
>> git bisect bad 27a4c827c34ac4256a190cc9d24607f953c1c459
>> # first bad commit: [27a4c827c34ac4256a190cc9d24607f953c1c459] fbcon: use the cursor blink interval provided by vt
>>
>> This seems unlikely, and I don't immediately see what is wrong, but reverting
>> the bad commit fixes the problem.
>
> No, that's correct, I have a fix coming soon to resolve this issue.
>

Not fixed as of next-20150527. Fix still in the works ?

Guenter