Skip to content

RFC scsi: mpt: Refactor and port to dma_* interface

Alex Dewar linux-scsi lkml 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 RFC Latest3 patches 3 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/3scsi: mpt: Remove unnecessary sleepFlag argument from functionsdrivers/message/fusion/mptbase.c drivers/message/fusion/mptbase.h drivers/message/fusion/mptctl.c +3 more+160 -2610 replies
2/3scsi: mpt: Replace a few uses of GFP_ATOMIC with GFP_KERNELdrivers/message/fusion/mptbase.c+4 -40 replies
3/3scsi: mpt: Port from pci_* to dma_* interfacedrivers/message/fusion/mptbase.c drivers/message/fusion/mptctl.c drivers/message/fusion/mptfc.c +2 more+248 -2120 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [PATCH RFC 0/3] scsi: mpt: Refactor and port to dma_* interface

Hi all,

I started porting over the mpt code from using the old pci_* compat functions to their dma_* counterparts, but realised that there are many functions which have a sleepFlags argument, which is actually almost never actually needed: almost all of the functions are always called with sleepFlag == CAN_SLEEP anyway. The first patch fixes these cases and could be applied by itself as a general tidy-up.

The other two patches are functional changes and so I added the RFC tag just to be extra cautious. Both of these patches involve changing some allocations from GFP_ATOMIC to GFP_KERNEL so I wanted to make sure that I wasn't introducing bugs. (Related question: Can you sleep in an ioctl context....?)

Any feedback would be greatly appreciated!

Best,
Alex