staging: zcache: delete it (96256460) · Commits · e / devices / android_kernel_fairphone_FP4

drivers/staging/Kconfig

+0 −2

Original line number	Diff line number	Diff line
		@@ -130,8 +130,6 @@ source "drivers/staging/sb105x/Kconfig"

		source "drivers/staging/fwserial/Kconfig"

		source "drivers/staging/zcache/Kconfig"

		source "drivers/staging/goldfish/Kconfig"

		source "drivers/staging/netlogic/Kconfig"

drivers/staging/Makefile

+0 −1

Original line number	Diff line number	Diff line
		@@ -58,7 +58,6 @@ obj-$(CONFIG_DRM_IMX) += imx-drm/
		obj-$(CONFIG_DGRP) += dgrp/
		obj-$(CONFIG_SB105X) += sb105x/
		obj-$(CONFIG_FIREWIRE_SERIAL) += fwserial/
		obj-$(CONFIG_ZCACHE) += zcache/
		obj-$(CONFIG_GOLDFISH) += goldfish/
		obj-$(CONFIG_USB_DWC2) += dwc2/
		obj-$(CONFIG_LUSTRE_FS) += lustre/

drivers/staging/zcache/Kconfig

deleted100644 → 0

+0 −59

Original line number	Diff line number	Diff line
		config ZCACHE
		tristate "Dynamic compression of swap pages and clean pagecache pages"
		depends on CRYPTO=y && SWAP=y && CLEANCACHE && FRONTSWAP
		select CRYPTO_LZO
		default n
		help
		Zcache doubles RAM efficiency while providing a significant
		performance boosts on many workloads. Zcache uses
		compression and an in-kernel implementation of transcendent
		memory to store clean page cache pages and swap in RAM,
		providing a noticeable reduction in disk I/O.

		config ZCACHE_DEBUG
		bool "Enable debug statistics"
		depends on DEBUG_FS && ZCACHE
		default n
		help
		This is used to provide an debugfs directory with counters of
		how zcache is doing. You probably want to set this to 'N'.

		config RAMSTER
		tristate "Cross-machine RAM capacity sharing, aka peer-to-peer tmem"
		depends on CONFIGFS_FS=y && SYSFS=y && !HIGHMEM && ZCACHE
		depends on NET
		# must ensure struct page is 8-byte aligned
		select HAVE_ALIGNED_STRUCT_PAGE if !64BIT
		default n
		help
		RAMster allows RAM on other machines in a cluster to be utilized
		dynamically and symmetrically instead of swapping to a local swap
		disk, thus improving performance on memory-constrained workloads
		while minimizing total RAM across the cluster. RAMster, like
		zcache2, compresses swap pages into local RAM, but then remotifies
		the compressed pages to another node in the RAMster cluster.

		config RAMSTER_DEBUG
		bool "Enable ramster debug statistics"
		depends on DEBUG_FS && RAMSTER
		default n
		help
		This is used to provide an debugfs directory with counters of
		how ramster is doing. You probably want to set this to 'N'.

		# Depends on not-yet-upstreamed mm patches to export end_swap_bio_write and
		# __add_to_swap_cache, and implement __swap_writepage (which is swap_writepage
		# without the frontswap call. When these are in-tree, the dependency on
		# BROKEN can be removed
		config ZCACHE_WRITEBACK
		bool "Allow compressed swap pages to be writtenback to swap disk"
		depends on ZCACHE=y && BROKEN
		default n
		help
		Zcache caches compressed swap pages (and other data) in RAM which
		often improves performance by avoiding I/O's due to swapping.
		In some workloads with very long-lived large processes, it can
		instead reduce performance. Writeback decompresses zcache-compressed
		pages (in LRU order) when under memory pressure and writes them to
		the backing swap disk to ameliorate this problem. Policy driving
		writeback is still under development.

drivers/staging/zcache/Makefile

deleted100644 → 0

+0 −8

Original line number	Diff line number	Diff line
		zcache-y := zcache-main.o tmem.o zbud.o
		zcache-$(CONFIG_ZCACHE_DEBUG) += debug.o
		zcache-$(CONFIG_RAMSTER_DEBUG) += ramster/debug.o
		zcache-$(CONFIG_RAMSTER) += ramster/ramster.o ramster/r2net.o
		zcache-$(CONFIG_RAMSTER) += ramster/nodemanager.o ramster/tcp.o
		zcache-$(CONFIG_RAMSTER) += ramster/heartbeat.o ramster/masklog.o

		obj-$(CONFIG_ZCACHE) += zcache.o

drivers/staging/zcache/TODO

deleted100644 → 0

+0 −64

Original line number	Diff line number	Diff line

		ZCACHE PLAN FOR PROMOTION FROM STAGING

		Last updated: Feb 13, 2013

		PLAN STEPS

		1. merge zcache and ramster to eliminate horrible code duplication
		2. converge on a predictable, writeback-capable allocator
		3. use debugfs instead of sysfs (per akpm feedback in 2011)
		4. zcache side of cleancache/mm WasActive patch
		5. zcache side of frontswap exclusive gets
		6. zcache must be able to writeback to physical swap disk
		(per Andrea Arcangeli feedback in 2011)
		7. implement adequate policy for writeback
		8. frontswap/cleancache work to allow zcache to be loaded
		as a module
		9. get core mm developer to review
		10. incorporate feedback from review
		11. get review/acks from 1-2 additional mm developers
		12. incorporate any feedback from additional mm reviews
		13. propose location/file-naming in mm tree
		14. repeat 9-13 as necessary until akpm is happy and merges

		STATUS/OWNERSHIP

		1. DONE as part of "new" zcache; in staging/zcache for 3.9
		2. DONE as part of "new" zcache (cf zbud.[ch]); in staging/zcache for 3.9
		(this was the core of the zcache1 vs zcache2 flail)
		3. DONE as part of "new" zcache; in staging/zcache for 3.9
		4. DONE (w/caveats) as part of "new" zcache; per cleancache performance
		feedback see https://lkml.org/lkml/2011/8/17/351, in
		staging/zcache for 3.9; dependent on proposed mm patch, see
		https://lkml.org/lkml/2012/1/25/300
		5. DONE as part of "new" zcache; performance tuning only,
		in staging/zcache for 3.9; dependent on frontswap patch
		merged in 3.7 (33c2a174)
		6. DONE (w/caveats), prototyped as part of "new" zcache, had
		bad memory leak; reimplemented to use sjennings clever tricks
		and proposed mm patches with new version in staging/zcache
		for 3.9, see https://lkml.org/lkml/2013/2/6/437;
		7. PROTOTYPED as part of "new" zcache; in staging/zcache for 3.9;
		needs more review (plan to discuss at LSF/MM 2013)
		9. IN PROGRESS; owned by Konrad Wilk; Mel Gorman provided
		great feedback in August 2012 (unfortunately of "old"
		zcache)
		11. NOT DONE; owned by Konrad Wilk and Bob Liu
		12. TBD (depends on quantity of feedback)
		13. PROPOSED; one suggestion proposed by Dan; needs more ideas/feedback
		14. TBD (depends on feedback)

		WHO NEEDS TO AGREE

		Not sure. Seth Jennings is now pursuing a separate but semi-parallel
		track. Akpm clearly has to approve for any mm merge to happen. Minchan
		Kim has interest but may be happy if/when zram is merged into mm. Konrad
		Wilk may be maintainer if akpm decides compression is maintainable
		separately from the rest of mm. (More LSF/MM 2013 discussion.)

		ZCACHE FUTURE NEW FUNCTIONALITY

		A. Support zsmalloc as an alternative high-density allocator
		(See https://lkml.org/lkml/2013/1/23/511)
		B. Possibly support three zbuds per pageframe when space allows