wireguard-openbsd/sbin/dump/traverse.c, branch jd/queueboosts

Cast a multiplicand to ino_t to avoid overflow in 32*32=64 calculations

2018-04-26T17:40:48Z

ok pirofti@ krw@

Adjust comments regarding use of use of MAXFRAG, or

2015-01-20T18:22:20Z

delete if now possible ok guenther

Replace with and other less dirty headers where

2015-01-16T06:39:28Z

possible. Annotate lines with their current reasons. Switch to PATH_MAX, NGROUPS_MAX, HOST_NAME_MAX+1, LOGIN_NAME_MAX, etc. Change MIN() and MAX() to local definitions of MINIMUM() and MAXIMUM() where sensible to avoid pulling in the pollution. These are the files confirmed through binary verification. ok guenther, millert, doug (helped with the verification protocol)

Add internal buffering when dumping a device with non-512 byte

2014-07-10T09:33:43Z

sectors. This ensures that the requested data, and all the requested data, is actually read from the device. Should have no effect on 512-byte sector devices, but makes dump work on 4096-byte sector devices. Problem noted by and fix tested by David Vasek.

Type cleanup:

2014-06-13T20:43:06Z

* Move all off_t variables that don't look like file sizes to int64_t. * Switch blockswritten to int64_t, so it won't wrap at 2TB. * Same for blocksthisvol (from deraadt@). * Switch xferrate (from tedu@) and blocksperfile from long to uint64_t. * Since blocksperfile can be set with -B, move numarg() from long to long long and don't mark small integer constant arguments as long. ok deraadt@, tedu@

Oops. We're reporting DEV_BSIZE values in error messages, so call

2014-05-31T14:15:22Z

them 'blocks' not 'sectors'.

Cast daddr_t variable (blkno) to (long long) when printing with %lld.

2014-05-31T11:28:44Z

Where trying to pread() a single disk sector, the i/o must be for the

2014-05-27T12:35:40Z

actual disk sector size and not DEV_BSIZE. The sector size must be obtained via the disklabel. Larger i/o's must be multiple sectors, so when retrying with a smaller size shrink the attempt by one sector and not DEV_BSIZE. Of course if your d_secsize is DEV_BSIZE, this will all be a no-op. This does not make non-512-byte sectors work, but puts in place the logic to get and use the disklabel info. Which makes the next diffs bite sized and focused. ok guenther@

Nuke last of the illusionary 'dev_bsize' and 'dev_bshift' variables in

2014-05-24T21:49:09Z

favour of DEV_BSIZE. No-op on 512-byte sector devices.

Nuke some trailing whitespace and break some long lines to shrink

2014-05-21T17:38:30Z

subsequent diffs.