diff options
author | zilio nicolas <nicolas.zilio@kit.edu> | 2015-08-26 14:45:18 +0200 |
---|---|---|
committer | zilio nicolas <nicolas.zilio@kit.edu> | 2015-08-26 14:45:18 +0200 |
commit | 1e5f0b6d5e02c0dc11bedefa92fc3c5bb406845a (patch) | |
tree | a978bd11f64f619ee8f40e42e73a582754c122de /docs/ToDo | |
parent | 0f298bd861ac8d847f33d6b8bc73b089d2749bbb (diff) | |
parent | 6bad94bb8546a3a5595d340e7a2d809635e3bd5d (diff) | |
download | pcitool-1e5f0b6d5e02c0dc11bedefa92fc3c5bb406845a.tar.gz pcitool-1e5f0b6d5e02c0dc11bedefa92fc3c5bb406845a.tar.bz2 pcitool-1e5f0b6d5e02c0dc11bedefa92fc3c5bb406845a.tar.xz pcitool-1e5f0b6d5e02c0dc11bedefa92fc3c5bb406845a.zip |
pull from server
Diffstat (limited to 'docs/ToDo')
-rw-r--r-- | docs/ToDo | 19 |
1 files changed, 10 insertions, 9 deletions
@@ -1,23 +1,24 @@ High Priority (we would need it for IPE Camera) ============= - 1. Serialize access to the registers across applications - 2. Protect kmem_entries in the driver using spinlock - 3. Protect mmap operation (multiple kernel calls) with some locking mechanism - 4. Allow overriding of registers and banks (performance?). + 1. Allow overriding of registers and banks (performance?). Normal Priority (it would make just few things a bit easier) =============== - 1. Implement software registers (stored in kernel-memory) - 2. Implement pcilib_configure_autotrigger - 3. Provide OR and AND operations on registers in cli - 4. Support writting a data from a binary file in cli + 1. Implement pcilib_configure_autotrigger + 2. Provide OR and AND operations on registers in cli + 3. Support writting a data from a binary file in cli Low Priority (only as generalization for other projects) ============ 1. XML configurations describing registers (and DMA engines?) - 2. Access register/bank lookups using hash tables + 2. Access register/bank/lock lookups using hash tables 3. Support for Network Registers and Network DMA 4. Define a syntax for register dependencies / delays (?) 5. Use pthread_condition_t instead of polling 6. Support FIFO reads/writes from/to registers + 7. We managed kmem performance using next kmem prediction, but it is still wise to provide additionally a binary tree for faster search +Performance +=========== + 1. Even with fully algined data, glibc defaults to __memcpy_sse2_unaligned (called from ipecamera_data_callback and fastwriter_push). Can we do something? +
\ No newline at end of file |