GitLab

Some boards feature non-removable memory devices with a no-fit option that
can only detected by seeing whether the device responds or not once we
start trying to talk to it. To support these, defer drive allocation until
the UnitAttached service call. This is already designed to dynamically
add extra drives in case a shared MMC bus is found, so the only difference
is that the initial number of drives is 0, not 1. Note that there is still
a minimum of 1 drive per slot for removable slots, since a removable slot
may be empty at module initialisation time, and there's no way to
electrically distinguish an empty socket from a no-fit socket.

Version 0.11. Tagged as 'SDFS-0_11'

Some controllers (such as the eMMC controller in the Pinebook) require
transfers to be limited to a smaller number of blocks per operation than is
permitted by a generic SDHCI controller.

Some genuinely dead code identified and removed, plus an out-of-bounds memory
access in case all possible drives had already been allocated. Some minor
printf format specifiers addressed (should have no effect) and removed useless
variable.

Detail:
* Uses CModule's ASM2TXT feature to simplify autogeneration of SDFSErr header.
* Fix case of one #include to suit case-sensitive filesystems.
* .gitignore populated with all temporary file paths.

Admin:
  Requires BuildSys-7_47

Version 0.10. Not tagged

Missing {DictTokens} added, otherwise when the CmdHelp gets concatenated it's never seen a directive and does nothing with the {Token0}.

Version 0.10. Tagged as 'SDFS-0_10'

SDFS_ReadCardInfo,1 allows the logical drive number to be mapped to bus/slot/RCA in a roughly analogous way to ADFS_IDEDeviceInfo. Otherwise, there's no way to map from drive number to the physical card.
Exports & docs updated.

Version 0.09. Tagged as 'SDFS-0_09'

Detail:
  On these boards, because there's no card-detect line, SDIODriver only
  issues the UnitAttached and UnitDetached service calls in its module
  init/final code and when one of the rescan SWIs are called (usually by
  SDFS). The "maybe-changed" timer is implemented within SDFS; when it has
  elapsed, any DiscOp will try to select a card with the last-known-good ID
  in that slot; if this fails (because there's no card, or the card has
  changed, or even if the same card has been removed and reinserted) SDFS
  will get SDIODriver to rescan the slot, at which point SDFS is re-entered
  via the UnitAttached service call, which resets the "maybe-changed" timer
  to 1 second.

  The problem was that SDFS reused the same function that handles
  Service_SDIOUnitAttached during its module initialisation, so for the
  first second of SDFS's life, it believed the card ID which it was
  informed of by SDIODriver is valid and has been accessed within the last
  second. Except that if the card has actually been removed and reinserted,
  the command issued by SDFS ends up timing out, resulting in disc error 08.
Admin:
  Tested on Pi B+ (assumed identical behaviour on Pi 2)

Version 0.08. Tagged as 'SDFS-0_08'

Detail:
  The eMMC chip on the Compute module is now treated as a hard drive by SDFS.
  A number of bugs (relating to top-bit-set disc addresses for example) were
  required to facilitate this. If a board has an eMMC chip, then the default
  SDFS drive is drive 4 rather than drive 0; this doesn't require any change
  to the default CMOS values in the kernel or HAL (SDFS adjusts its
  interpretation of the CMOS byte instead).
Admin:
  Tested on a Compute module (of course).

Version 0.07. Tagged as 'SDFS-0_07'

Detail:
  Non-removable SDIO devices were never supposed to be allocated an SDFS
  drive, but were by accident in the secondary (UnitAttached rather than
  SlotAttached) service call.
Admin:
  Applies to boards such as the IGEPv2, with SDIO WiFi chips on the board.


Version 0.06. Tagged as 'SDFS-0_06'

Detail:
  For removable slots without a card detect pin, use an alternate state
  machine based upon timeouts to detect card removal or card changes. Note
  that this requires cooperation from FileCore (the same code paths as with
  selected floppy drives that had faulty disc changed pins) to avoid data
  loss by assuming that a disc might need remounting if a certain time has
  elapsed since the last successful operation. To avoid a significant speed
  penalty from deselecting and reselecting cards, remounting isn't used for
  rapid back-to-back operations.
Admin:
  Already in use in RC12a.

Version 0.05. Tagged as 'SDFS-0_05'

Version 0.04. Tagged as 'SDFS-0_04'

Detail:
  c/module - A typo in the startup code resulted in the service_SDIOSlotAttached 'integrated' parameter being calculated incorrectly.
Admin:
  Tested on Pandora
  2nd SD slot now appears on iconbar if slot is empty on boot (previously would only appear if slot occupied)


Version 0.03. Tagged as 'SDFS-0_03'

Detail:
  MiscOp 0 turns out to pass the drive number *both* in R1 and in the top
  3 bits of R2. SDFS was failing to mask the drive bits out of R2, which
  caused some (but not all) cards to generate an error. Obviously, this
  didn't affect drive 0.
Admin:
  Tested on Beagleboard "Zippy" expansion card

Version 0.02. Tagged as 'SDFS-0_02'