On 26.04.2011 05:03, Bean wrote: > Hi, > > It's not decided by fs driver, it's configured by a flag in > grub_file_open, which is passed to the disk driver. There is only a > few places where we can expect large file, such as linux/initrd > command. Yet worse, such a high level shouldn't care about low-level at all. > And in this case, caching is useless since we only use the > data once and caching would only flush out useful data unnecessarily. > By setting a flag to indicate direct read is required, we can optimize > access for such situation while keeping the cache for others. This is > similar to the pass through flag for linux/windows. I don't think that the intent is the same. Under OS considerations are more along transaction-safeness and barriers for databases and cache disabling for system software. Normal application shouldn't use those flags. >> Which is an example of bad design. Rather than improving the existing >> function to do both caching and unbroken read (like in my 4096 branch) >> you have 2 functions and force upper layers to do the tradeoffs and care >> about matters which should be abstracted and invisible to them. > _______________________________________________ > Grub-devel mailing list > Grub-devel@gnu.org > https://lists.gnu.org/mailman/listinfo/grub-devel >
-- Regards Vladimir 'φ-coder/phcoder' Serbinenko
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Grub-devel mailing list Grub-devel@gnu.org https://lists.gnu.org/mailman/listinfo/grub-devel
