cvs commit: src/sys/dev/ata ata-all.c ata-all.h ata-card.c ata-cbus.c ata-chipset.c ata-disk.c ata-dma.c ata-isa.c ata-lowlevel.c ata-pci.c ata-queue.c ata-raid.c atapi-cam.c atapi-cd.c atapi-fd.c atapi-tape.c

[Søren Schmidt]

sos         2004/01/14 13:26:35 PST

  FreeBSD src repository

  Modified files:
    sys/dev/ata          ata-all.c ata-all.h ata-card.c ata-cbus.c 
                         ata-chipset.c ata-disk.c ata-dma.c 
                         ata-isa.c ata-lowlevel.c ata-pci.c 
                         ata-queue.c ata-raid.c atapi-cam.c 
                         atapi-cd.c atapi-fd.c atapi-tape.c 
  Log:
  Use UMA instead of plain malloc for getting ATA request storage.
  This gives +10% performance on simple tests, so definitly worth it.
  A few percent more could be had by not using M_ZERO'd alloc's, but
  we then need to clear fields all over the place to be safe, and
  that was deemed not worth the trouble (and it makes life dangerous).
  
  Revision  Changes    Path
  1.201     +7 -2      src/sys/dev/ata/ata-all.c
  1.70      +5 -3      src/sys/dev/ata/ata-all.h
  1.22      +1 -0      src/sys/dev/ata/ata-card.c
  1.10      +1 -0      src/sys/dev/ata/ata-cbus.c
  1.56      +1 -0      src/sys/dev/ata/ata-chipset.c
  1.167     +1 -2      src/sys/dev/ata/ata-disk.c
  1.125     +1 -0      src/sys/dev/ata/ata-dma.c
  1.20      +1 -0      src/sys/dev/ata/ata-isa.c
  1.27      +1 -0      src/sys/dev/ata/ata-lowlevel.c
  1.74      +1 -0      src/sys/dev/ata/ata-pci.c
  1.16      +6 -27     src/sys/dev/ata/ata-queue.c
  1.75      +1 -0      src/sys/dev/ata/ata-raid.c
  1.31      +1 -0      src/sys/dev/ata/atapi-cam.c
  1.160     +2 -2      src/sys/dev/ata/atapi-cd.c
  1.92      +1 -0      src/sys/dev/ata/atapi-fd.c
  1.87      +1 -0      src/sys/dev/ata/atapi-tape.c