Akira Hayakawa

@bash99 How do you write back dirty caches with wb? and how do you know its completion?

@bash99 As the baseline you need the result with HDD only. I think dmwb doesn't do any performance gain with the workload. Usually, client application doesn't write in such a...

is it possible to limit the IO range by "--size=4G"?

I have done the similar experiment before (this test isn't workable now because it's not maintained) ```scala test("writeback sorting effect") { val amount = 128 1 slowDevice(Sector.G(2)) { backing =>...

@bash99 > As Eio has known bugs when fsck back device when has dirty data in cache device, they maybe do some unsafe optimized That must be so but it's...

The test in writeboost-test-suite uses 0 as the offset. `s"0 ${backingDev.bdev.size} writeboost ${backingDev.bdev.path} ${cacheDev.bdev.path}" + optionalArgs` but to change this number will cause wide range of changes in other codebase....

The motivation to use offset other than 0 is creating a large virtual device that is concatenating multiple virtual devices or mappings. When implementing linear target, it sounds essential but...

@onlyjob Sharing a caching device with two dm-writeboost'd devices is invalid. I once considered this by embedding some unique identifier of the backing device into the super block of the...

@onlyjob I wrote a test to reproduce the error message you saw in dmesg ``` scala class REPRO_115 extends DMTestSuite { test("endless loop when try to use active caching device")...

I wrote a test for the case but there is no error. ``` scala test("making another device while in use") { Memory(Sector.M(8)) { caching => Memory(Sector.M(64)) { backing1 => Memory(Sector.M(64))...