The block cache implementation in loader has proven to be almost useless, and in worst case even slowing down the disk reads due to insufficient cache size and extra memory copy.
Also the current cache implementation does not cache reads from CDs, or work with zfs built on top of multiple disks.
Instead of an LRU, this code uses a simple hash (O(1) read from cache), and instead of a single global cache, a separate cache per block device.
The cache also implements limited read-ahead to increase performance.
To simplify read ahead management, the read ahead will not wrap over bcache end, so in worst case, single block physical read will be performed to fill the last block in bcache.
Booting from a virtual CD over IPMI:
0ms latency, before: 27 second, after: 7 seconds
60ms latency, before: over 12 minutes, after: under 5 minutes.
Submitted by: Toomas Soome <tsoome@me.com>
Reviewed by: delphij (previous version), emaste (previous version)
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D4713
only happen on every Nth call. Update the existing twiddle() calls done in
various IO loops to roughly reflect the relative IO sizes. That is, tftp
and nfs call twiddle() on every 1K block, ufs on every filesystem block,
so the network calls now use a much larger divisor than disk IO calls.
Also add a new twiddle_divisor() function that allows an application to set
a global divisor that is applied on top of the per-call divisors. Nothing
calls this yet, but loader(8) will be using it to further throttle the
cursor for slow serial consoles.
Based on r134760:
Reset the seek pointer to 0 when a file is successfully opened,
since otherwise the initial seek offset will contain the directory
offset of the filesystem block that contained its directory entry.
This bug was mostly harmless because typically the directory is
less than one filesystem block in size so the offset would be zero.
It did however generally break loading a kernel from the (large)
kernel compile directory.
Also reset the seek pointer when a new inode is opened in read_inode(),
though this is not actually necessary now because all callers set
it afterwards.
PR: 177328
Submitted by: Eric van Gyzen
Reviewed by: iedowse
MFC after: 5 days
- bzipfs and gzipfs now properly return errno values directly from their
read routines rather than returning -1.
- missing errno values on error returns for the seek routines on almost
all filesystems were added.
- fstat() now returns -1 if an error occurs rather than ignoring it.
- nfs's readdir() routine now reports valid errno values if an error or
EOF occurs rather than EPERM (It was just returning 0 for success and
1 for failure).
- nullfs used the wrong semantics for every function besides close() and
seek(). Getting it right for close() appears to be an accident at that.
- read() for buffered files no longer returns 0 (EOF) if an error occurs,
but returns -1 instead.
