Re: [axboe-block:for-next] [block] bd4a633b6f: fsmark.files_per_sec -64.5% regression

Christoph Hellwig <[email protected]> Mon, 24 Jun 2024 17:20:28 +0200
Newsgroups dev.linux.lists.drbd-dev,dev.linux.lists.dm-devel,dev.linux.lists.nvdimm,dev.linux.lists.oe-lkp,dev.linux.lists.virtualization,org.infradead.lists.linux-mtd,org.infradead.lists.linux-nvme,org.infradead.lists.linux-um,org.kernel.vger.ceph-devel,org.kernel.vger.linux-bcache,org.kernel.vger.linux-block,org.kernel.vger.linux-kernel,org.kernel.vger.linux-m68k,org.kernel.vger.linux-mmc,org.kernel.vger.linux-raid,org.kernel.vger.linux-s390,org.kernel.vger.linux-scsi,org.ozlabs.lists.linuxppc-dev,org.xenproject.lists.xen-devel
Message-ID <[email protected]> 
On Mon, Jun 24, 2024 at 03:45:57PM +0200, Niklas Cassel wrote:
> Seems to be ATA SSD:
> https://download.01.org/0day-ci/archive/20240624/[email protected]/job.yaml
> 
> ssd_partitions: "/dev/disk/by-id/ata-INTEL_SSDSC2BG012T4_BTHC428201ZX1P2OGN-part1"
> 
> Most likely btrfs does something different depending on the nonrot flag
> being set or not. (And like you are suggesting, most likely the value of
> the nonrot flag is somehow different after commit bd4a633b6f)

Yes, btrfs does.  That's why I'm curious about the before and after,
as I can't see any way how they would be set differently.  Right now
I can only claim with vitual AHCI devices, which claim to be rotational,
though.