On Wed, Oct 20, 2021 at 11:11 PM Scargall, Steve <[email protected]> wrote: > > Hi Adam, > > This is likely related to an issue that was reported to the Linux NVDIMM > email list > (https://lore.kernel.org/linux-block/CAHj4cs87BapQJcV0a=M6=dc9prsgh6qzqjet9fbjlk1ashn...@mail.gmail.com/) > This issue was fixed by this patchset:https://lore.kernel.org/nvdimm/[email protected]/
And they are still in the review process. > So the bisecting shows it was introduced with below commit: > > commit 8e141f9eb803e209714a80aa6ec073893f94c526 > Author: Christoph Hellwig <[email protected]> > Date: Wed Sep 29 09:12:40 2021 +0200 > > block: drain file system I/O on del_gendisk > > /Steve > > -----Original Message----- > From: Adam Borowski <[email protected]> > Sent: Wednesday, October 20, 2021 8:04 AM > To: [email protected]; Williams, Dan J <[email protected]>; > Verma, Vishal L <[email protected]> > Subject: Re: ndctl hangs with big memmap=! fakepmem > > On Wed, Oct 20, 2021 at 03:23:08PM +0200, Adam Borowski wrote: > > Hi! > > After bumping fakepmem sizes from 4G!20G 4G!36G to 32G!20G 32G!192G, > > ndctl hangs. Eg, at boot: > > > > [ 725.642546] INFO: task ndctl:2486 blocked for more than 604 seconds. > > [ 725.649586] Not tainted 5.15.0-rc6-vanilla-00020-gd9abdee5fd5a #1 > > > [ 725.677539] ? __schedule+0x30b/0x14e0 [ 725.681975] ? > > kernfs_put.part.0+0xd4/0x1a0 [ 725.686841] ? > > kmem_cache_free+0x28b/0x2b0 [ 725.691622] ? schedule+0x44/0xb0 [ > > 725.695622] ? blk_mq_freeze_queue_wait+0x62/0x90 > > [ 725.701009] ? do_wait_intr_irq+0xc0/0xc0 [ 725.705703] ? > > del_gendisk+0xcf/0x220 [ 725.710050] ? release_nodes+0x38/0xa0 > > On 5.14.14 all is fine. Should I bisect? > > > Meow! > -- > ⢀⣴⠾⠻⢶⣦⠀ > ⣾⠁⢠⠒⠀⣿⡁ Remember, the S in "IoT" stands for Security, while P stands ⢿⡄⠘⠷⠚⠋⠀ > for Privacy. > ⠈⠳⣄⠀⠀⠀⠀ > -- Best Regards, Yi Zhang
