| Message ID | 20180609225511.24890-1-aford173@gmail.com |
|---|---|
| State | Superseded |
| Delegated to: | Tom Rini |
| Headers | show |
| Series | [U-Boot] block: Add SPL_BLOCK_CACHE and default n | expand |
On Sat, Jun 9, 2018 at 11:55 PM Adam Ford <aford173@gmail.com> wrote: > > When enabling BLOCK_CACHE on devices with limited RAM during SPL, > some devices may not boot. This creates an option to enable > block caching in SPL by defaults off. It is dependent on BLOCK_CACHE > and SPL_BLK > > Fixes: 46960ad6d09b ("block: Have BLOCK_CACHE default to y in some cases") > > Signed-off-by: Adam Ford <aford173@gmail.com> > -- > > V2: Clean up macros and ifdefs > > diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig > index 0792373cfc..8ef363b3d4 100644 > --- a/drivers/block/Kconfig > +++ b/drivers/block/Kconfig > @@ -37,6 +37,13 @@ config BLOCK_CACHE > it will prevent repeated reads from directory structures and other > filesystem data structures. > > +config SPL_BLOCK_CACHE > + bool "Use block device cache in SPL" > + depends on BLOCK_CACHE && SPL_BLK Seems like an odd dependency - to use the block cache in SPL, you also have to have block cache in full U-Boot? I'm guessing `depends on SPL_BLK` would be sufficient > + default n > + help > + This option enables the disk-block cache in SPL > + > config IDE > bool "Support IDE controllers" > select HAVE_BLOCK_DEVICE > diff --git a/drivers/block/Makefile b/drivers/block/Makefile > index 5fcafb193e..0e80ce9405 100644 > --- a/drivers/block/Makefile > +++ b/drivers/block/Makefile > @@ -11,4 +11,4 @@ endif > > obj-$(CONFIG_IDE) += ide.o > obj-$(CONFIG_SANDBOX) += sandbox.o > -obj-$(CONFIG_BLOCK_CACHE) += blkcache.o > +obj-$(CONFIG_$(SPL_)BLOCK_CACHE) += blkcache.o > diff --git a/include/blk.h b/include/blk.h > index fc0c239e46..bd90e39932 100644 > --- a/include/blk.h > +++ b/include/blk.h > @@ -111,7 +111,8 @@ struct blk_desc { > #define PAD_TO_BLOCKSIZE(size, blk_desc) \ > (PAD_SIZE(size, blk_desc->blksz)) > > -#ifdef CONFIG_BLOCK_CACHE > +#if (CONFIG_IS_ENABLED(BLOCK_CACHE) && !CONFIG_IS_ENABLED(SPL_BUILD)) || \ > + (CONFIG_IS_ENABLED(SPL_BLOCK_CACHE) && CONFIG_IS_ENABLED(SPL_BUILD)) Assuming I read it right, you shouldn't need all the &&/|| gyrations, just: #if CONFIG_IS_ENABLED(BLOCK_CACHE) Should get you what you want. > /** > * blkcache_read() - attempt to read a set of blocks from cache > * > -- > 2.17.1 > -- Alex Kiernan
On Sat, Jun 09, 2018 at 05:55:11PM -0500, Adam Ford wrote: > When enabling BLOCK_CACHE on devices with limited RAM during SPL, > some devices may not boot. This creates an option to enable > block caching in SPL by defaults off. It is dependent on BLOCK_CACHE > and SPL_BLK > > Fixes: 46960ad6d09b ("block: Have BLOCK_CACHE default to y in some cases") > > Signed-off-by: Adam Ford <aford173@gmail.com> > -- > > V2: Clean up macros and ifdefs [snip] > diff --git a/include/blk.h b/include/blk.h > index fc0c239e46..bd90e39932 100644 > --- a/include/blk.h > +++ b/include/blk.h > @@ -111,7 +111,8 @@ struct blk_desc { > #define PAD_TO_BLOCKSIZE(size, blk_desc) \ > (PAD_SIZE(size, blk_desc->blksz)) > > -#ifdef CONFIG_BLOCK_CACHE > +#if (CONFIG_IS_ENABLED(BLOCK_CACHE) && !CONFIG_IS_ENABLED(SPL_BUILD)) || \ > + (CONFIG_IS_ENABLED(SPL_BLOCK_CACHE) && CONFIG_IS_ENABLED(SPL_BUILD)) > /** > * blkcache_read() - attempt to read a set of blocks from cache > * CONFIG_IS_ENABLED(FOO) has the logic for checking for SPL_FOO and SPL_BUILD and so forth. Thanks!
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig index 0792373cfc..8ef363b3d4 100644 --- a/drivers/block/Kconfig +++ b/drivers/block/Kconfig @@ -37,6 +37,13 @@ config BLOCK_CACHE it will prevent repeated reads from directory structures and other filesystem data structures. +config SPL_BLOCK_CACHE + bool "Use block device cache in SPL" + depends on BLOCK_CACHE && SPL_BLK + default n + help + This option enables the disk-block cache in SPL + config IDE bool "Support IDE controllers" select HAVE_BLOCK_DEVICE diff --git a/drivers/block/Makefile b/drivers/block/Makefile index 5fcafb193e..0e80ce9405 100644 --- a/drivers/block/Makefile +++ b/drivers/block/Makefile @@ -11,4 +11,4 @@ endif obj-$(CONFIG_IDE) += ide.o obj-$(CONFIG_SANDBOX) += sandbox.o -obj-$(CONFIG_BLOCK_CACHE) += blkcache.o +obj-$(CONFIG_$(SPL_)BLOCK_CACHE) += blkcache.o diff --git a/include/blk.h b/include/blk.h index fc0c239e46..bd90e39932 100644 --- a/include/blk.h +++ b/include/blk.h @@ -111,7 +111,8 @@ struct blk_desc { #define PAD_TO_BLOCKSIZE(size, blk_desc) \ (PAD_SIZE(size, blk_desc->blksz)) -#ifdef CONFIG_BLOCK_CACHE +#if (CONFIG_IS_ENABLED(BLOCK_CACHE) && !CONFIG_IS_ENABLED(SPL_BUILD)) || \ + (CONFIG_IS_ENABLED(SPL_BLOCK_CACHE) && CONFIG_IS_ENABLED(SPL_BUILD)) /** * blkcache_read() - attempt to read a set of blocks from cache *
When enabling BLOCK_CACHE on devices with limited RAM during SPL, some devices may not boot. This creates an option to enable block caching in SPL by defaults off. It is dependent on BLOCK_CACHE and SPL_BLK Fixes: 46960ad6d09b ("block: Have BLOCK_CACHE default to y in some cases") Signed-off-by: Adam Ford <aford173@gmail.com> -- V2: Clean up macros and ifdefs