FINAL(?) SCSI tape mods: fix bsr/fsr in light of 01 Apr 2006 w32stape fixes for utilities, tweak to auto-scsi-mount logic to eliminate(?) WIN32 drive query inefficiencies

git-svn-id: file:///home/jj/hercules.svn/trunk@3802 956126f8-22a0-4046-8f4a-272fa8102e63
This commit is contained in:
Fish (David B Trout)
2006-04-05 22:34:15 +00:00
parent d4bfb10705
commit feff40c835
5 changed files with 178 additions and 182 deletions

View File

@@ -33,12 +33,25 @@ extern int fsf_scsitape ( DEVBLK *dev, BYTE *u
extern int bsf_scsitape ( DEVBLK *dev, BYTE *unitstat, BYTE code );
extern int rewind_scsitape ( DEVBLK *dev, BYTE *unitstat, BYTE code );
extern void rewind_unload_scsitape ( DEVBLK *dev, BYTE *unitstat, BYTE code );
extern int driveready_scsitape ( DEVBLK *dev, BYTE *unitstat, BYTE code );
extern int is_tape_mounted_scsitape ( DEVBLK *dev, BYTE *unitstat, BYTE code );
extern int finish_scsitape_open ( DEVBLK *dev, BYTE *unitstat, BYTE code );
extern void update_status_scsitape ( DEVBLK *dev );
extern void update_status_scsitape ( DEVBLK *dev, int mountstat_only );
extern void *scsi_tapemountmon_thread ( void *devblk );
extern void kill_stape_status_thread ( DEVBLK *dev );
// PROGRAMMING NOTE: I'm not sure of what the the actual/proper value
// should be (or is) for the following value but I've coded what seems
// to me to be a reasonable value for it. As you can probably guess
// based on its [admittedly rather verbose] name, it's the maximum
// amount of time that a SCSI tape drive query call should take (i.e.
// asking the system for the drive's status shouldn't, under normal
// circumstances, take any longer than this time). It should be set
// to the most pessimistic value we can reasonably stand, and should
// probably be at least as long as the host operating system's thread
// scheduling time-slice quantum. - Fish, April 2006
#define MAX_NORMAL_SCSI_DRIVE_QUERY_RESPONSE_TIMEOUT_USECS (25*1000)
#endif // defined(OPTION_SCSI_TAPE)
#endif // _SCSITAPE_H_