Florent Kermarrec [Mon, 13 Apr 2015 09:23:27 +0000 (11:23 +0200)]
liteeth: pep8 (E265)
Florent Kermarrec [Mon, 13 Apr 2015 09:07:50 +0000 (11:07 +0200)]
liteeth: pep8 (E261, E271)
Florent Kermarrec [Mon, 13 Apr 2015 08:58:45 +0000 (10:58 +0200)]
liteeth: pep8 (W292)
Florent Kermarrec [Mon, 13 Apr 2015 08:56:18 +0000 (10:56 +0200)]
liteeth: pep8 (E225)
Florent Kermarrec [Mon, 13 Apr 2015 08:48:59 +0000 (10:48 +0200)]
liteeth: pep8 (E222)
Florent Kermarrec [Mon, 13 Apr 2015 08:45:09 +0000 (10:45 +0200)]
liteeth: pep8 (E401)
Florent Kermarrec [Mon, 13 Apr 2015 08:39:46 +0000 (10:39 +0200)]
liteeth: pep8 (E203)
Florent Kermarrec [Mon, 13 Apr 2015 08:31:18 +0000 (10:31 +0200)]
liteeth: pep8 (E231)
Florent Kermarrec [Mon, 13 Apr 2015 08:23:33 +0000 (10:23 +0200)]
liteeth: pep8 (E201)
Florent Kermarrec [Mon, 13 Apr 2015 08:20:02 +0000 (10:20 +0200)]
liteeth: pep8 (E302)
Florent Kermarrec [Mon, 13 Apr 2015 07:53:43 +0000 (09:53 +0200)]
liteeth: pep8 (replace tabs with spaces)
Florent Kermarrec [Sun, 12 Apr 2015 20:09:46 +0000 (22:09 +0200)]
liteeth/phy/gmii: fix clock generation for mii mode (clock_pads.tx is an input)
Florent Kermarrec [Sun, 12 Apr 2015 18:58:23 +0000 (20:58 +0200)]
microudp.c: add #ifdef on ethmode (bios generation for gmii or mii was broken)
Florent Kermarrec [Sun, 12 Apr 2015 18:19:32 +0000 (20:19 +0200)]
liteeth/phy/gmii_mii: add pads registers in RX
Florent Kermarrec [Sun, 12 Apr 2015 18:11:08 +0000 (20:11 +0200)]
liteeth/phy/gmii_mii: avoid doubling pads register on TX
Florent Kermarrec [Sun, 12 Apr 2015 16:56:46 +0000 (18:56 +0200)]
liteeth/phy/__init__.py: add more comments
Florent Kermarrec [Sun, 12 Apr 2015 16:52:35 +0000 (18:52 +0200)]
microudp: fix if (
Florent Kermarrec [Sun, 12 Apr 2015 15:33:38 +0000 (17:33 +0200)]
liteeth/phy/gmii_mii: add clock counter and use it in bios to select mode
Florent Kermarrec [Sun, 12 Apr 2015 14:49:39 +0000 (16:49 +0200)]
liteeth/phy: add GMII/MII phy
for now swicth is manual, we will need a clk counter to allow software or logic to automatically switch between GMII and MII
Florent Kermarrec [Sun, 12 Apr 2015 13:50:20 +0000 (15:50 +0200)]
liteeth/phy/mii: simplify LiteEthPHYMIIRX using Converter
Florent Kermarrec [Sun, 12 Apr 2015 13:21:58 +0000 (15:21 +0200)]
liteeth/phy/mii: simplify LiteEthPHYMIITX using Converter
Florent Kermarrec [Sun, 12 Apr 2015 12:43:35 +0000 (14:43 +0200)]
liteeth/phy/mii: assign tx_er only if exists (as it's done on GMII)
Florent Kermarrec [Sun, 12 Apr 2015 12:27:29 +0000 (14:27 +0200)]
liteeth/phy/mii: allow use of MII phy on GMII/MII chips that do not have phy clock provided by the FPGA (tested on KC705)
Florent Kermarrec [Fri, 10 Apr 2015 16:57:06 +0000 (18:57 +0200)]
timer: revert prescaler (we will in fact use a software prescaler for uIP)
Florent Kermarrec [Fri, 10 Apr 2015 11:58:44 +0000 (13:58 +0200)]
timer: add prescaler
Robert Jordens [Fri, 10 Apr 2015 01:17:02 +0000 (19:17 -0600)]
s6ddrphy: redo phase_sel, get rid of CLOCK_DEDICATED_ROUTE
Sebastien Bourdeauducq [Fri, 10 Apr 2015 03:42:25 +0000 (11:42 +0800)]
liteeth: adapt to new ModuleTransformer
Florent Kermarrec [Wed, 8 Apr 2015 22:00:25 +0000 (00:00 +0200)]
litesata: update build core target generation
Florent Kermarrec [Wed, 8 Apr 2015 21:27:22 +0000 (23:27 +0200)]
lite*: finish ModuleTransformer adaptations (need to be tested on board)
Sebastien Bourdeauducq [Wed, 8 Apr 2015 16:34:36 +0000 (00:34 +0800)]
soc,cpuif: support user defined constants
Sebastien Bourdeauducq [Mon, 6 Apr 2015 15:53:04 +0000 (23:53 +0800)]
make: add target in build names
Sebastien Bourdeauducq [Mon, 6 Apr 2015 15:52:34 +0000 (23:52 +0800)]
soc: use new ModuleTransformer API
Robert Jordens [Thu, 2 Apr 2015 20:28:42 +0000 (14:28 -0600)]
lite*: adapt to new ModuleTransformer semantics
NOTE: There is loads of duplicated code between the lite*
modules that should be shared.
Florent Kermarrec [Fri, 3 Apr 2015 11:57:37 +0000 (13:57 +0200)]
soc/cpuif: fix CSR base generation for memories (name is already fullname)
Florent Kermarrec [Fri, 3 Apr 2015 10:45:32 +0000 (12:45 +0200)]
soc: add memory.name_override to name when adding csrbankarray.srams to csr_regions
Sebastien Bourdeauducq [Fri, 3 Apr 2015 09:43:46 +0000 (17:43 +0800)]
use str.format
Sebastien Bourdeauducq [Fri, 3 Apr 2015 09:43:29 +0000 (17:43 +0800)]
software/common.mak: fix alignment in quiet output
Florent Kermarrec [Fri, 3 Apr 2015 09:14:28 +0000 (11:14 +0200)]
soc/cpuif: fix get_csr_header when obj is Memory (thanks ccube)
Sebastien Bourdeauducq [Fri, 3 Apr 2015 08:00:07 +0000 (16:00 +0800)]
make.py: use os.path.join
Sebastien Bourdeauducq [Fri, 3 Apr 2015 05:23:28 +0000 (13:23 +0800)]
crt0-or1k: clean up indentation
Florent Kermarrec [Thu, 2 Apr 2015 10:18:43 +0000 (12:18 +0200)]
remove use of _r prefix on CSRs
Sebastien Bourdeauducq [Thu, 2 Apr 2015 09:17:33 +0000 (17:17 +0800)]
move gpio from cpu.peripherals to com
Sebastien Bourdeauducq [Thu, 2 Apr 2015 08:47:03 +0000 (16:47 +0800)]
libbase: implement flush_l2_cache for or1k
Sebastien Bourdeauducq [Thu, 2 Apr 2015 06:40:29 +0000 (14:40 +0800)]
minor cleanups
Sebastien Bourdeauducq [Thu, 2 Apr 2015 02:14:24 +0000 (10:14 +0800)]
Merge branch 'master' of github.com:m-labs/misoc
Florent Kermarrec [Wed, 1 Apr 2015 20:52:19 +0000 (22:52 +0200)]
adapt LiteSATA to new SoC
Florent Kermarrec [Wed, 1 Apr 2015 20:50:29 +0000 (22:50 +0200)]
adapt LiteEth to new SoC
Florent Kermarrec [Wed, 1 Apr 2015 20:45:57 +0000 (22:45 +0200)]
adapt LiteScope to new SoC
Florent Kermarrec [Wed, 1 Apr 2015 20:38:04 +0000 (22:38 +0200)]
soc/sdram: fix do_finalize
Sebastien Bourdeauducq [Wed, 1 Apr 2015 16:14:56 +0000 (00:14 +0800)]
soc: use set
Sebastien Bourdeauducq [Wed, 1 Apr 2015 16:09:38 +0000 (00:09 +0800)]
soc: simplify integrated memory parameters
Sebastien Bourdeauducq [Wed, 1 Apr 2015 15:41:55 +0000 (23:41 +0800)]
soc/sdram: minor cleanup
Sebastien Bourdeauducq [Wed, 1 Apr 2015 09:37:53 +0000 (17:37 +0800)]
litesata: adapt to new SoC API
Sebastien Bourdeauducq [Wed, 1 Apr 2015 09:32:45 +0000 (17:32 +0800)]
soc: remove cpu_boot_file argument
Sebastien Bourdeauducq [Wed, 1 Apr 2015 09:29:51 +0000 (17:29 +0800)]
soc: remove cpu_or_bridge and with_cpu arguments
Sebastien Bourdeauducq [Wed, 1 Apr 2015 08:49:32 +0000 (16:49 +0800)]
soc: retrieve csr and memory regions using methods
Sebastien Bourdeauducq [Wed, 1 Apr 2015 07:56:51 +0000 (15:56 +0800)]
soc: use add_wb_master function
Sebastien Bourdeauducq [Wed, 1 Apr 2015 07:48:56 +0000 (15:48 +0800)]
soc: simplify/fix csr busword
Sebastien Bourdeauducq [Wed, 1 Apr 2015 07:15:09 +0000 (15:15 +0800)]
soc: remove unnecessary imports
Sebastien Bourdeauducq [Wed, 1 Apr 2015 07:14:02 +0000 (15:14 +0800)]
soc: improve memory region conflict check
Sebastien Bourdeauducq [Wed, 1 Apr 2015 06:33:12 +0000 (14:33 +0800)]
soc: remove ns function
Florent Kermarrec [Sun, 29 Mar 2015 10:34:40 +0000 (12:34 +0200)]
sdram: remove redundant with_l2 parameter (equivalent to l2_size != 0)
Florent Kermarrec [Sat, 28 Mar 2015 22:18:08 +0000 (23:18 +0100)]
soc: limit main_ram_size to 256MB (we should modify mem_map to allow larger memories, this was the probably ARTIQ runtime issue....!!)
Florent Kermarrec [Sat, 28 Mar 2015 22:10:33 +0000 (23:10 +0100)]
soc: simplify main_ram_size computation and share it between LASMIcon and Minicon
Florent Kermarrec [Sat, 28 Mar 2015 15:35:15 +0000 (16:35 +0100)]
sdram/module: fix MT8JTF12864, rowbits is 14 and not 16.... (16 was used from the beginning, but it does not fix the runtime issue)
Florent Kermarrec [Sat, 28 Mar 2015 00:59:55 +0000 (01:59 +0100)]
sdram/phy/simphy: OK with DDR3
Florent Kermarrec [Sat, 28 Mar 2015 00:18:35 +0000 (01:18 +0100)]
sdram/phy/simphy: expose settings to user and test with DDR/LPDDR/DDR2
Florent Kermarrec [Sat, 28 Mar 2015 00:17:50 +0000 (01:17 +0100)]
sdram/core/lasmicon: add enabled parameter to refresher (for some simulations we need to disable it)
Florent Kermarrec [Sat, 28 Mar 2015 00:09:21 +0000 (01:09 +0100)]
sdram/module: clean up tREFI. (use 64ms/8k or 4k)
Sebastien Bourdeauducq [Fri, 27 Mar 2015 18:22:29 +0000 (19:22 +0100)]
Merge branch 'master' of https://github.com/m-labs/misoc
Robert Jordens [Thu, 26 Mar 2015 20:12:35 +0000 (14:12 -0600)]
pipistrello: add por reset counter
* this is a temporary fix that should be removed once the
combination of bitstream-in-flash, mor1kx, bios-in-flash works
Florent Kermarrec [Fri, 27 Mar 2015 17:24:19 +0000 (18:24 +0100)]
software/bios/sdram: small clean up
Florent Kermarrec [Fri, 27 Mar 2015 15:43:22 +0000 (16:43 +0100)]
software/bios/sdram: for now desactivate random on address test since it seems to trigger a L2 cache or LASMIcon bug on at least de0nano/minispartan6
Memtest sometimes reports 1 or 2 errors with de0nano/minispartan6 on this new test when used with LASMICON. Minicon seems fine. We will have to investigate on this issue.
Florent Kermarrec [Fri, 27 Mar 2015 14:49:16 +0000 (15:49 +0100)]
software/bios/sdram: add random addressing to memtest
testing memories with linear access is not good enough. Adding random addressing allow us to detect more eventual issues on our L2 cache or SDRAM controller.
Florent Kermarrec [Thu, 26 Mar 2015 22:45:35 +0000 (23:45 +0100)]
targets: revert use of integers in clocks/timings
Florent Kermarrec [Thu, 26 Mar 2015 22:27:37 +0000 (23:27 +0100)]
sdram: remove nbits from modules and databits from GeomSettings
Florent Kermarrec [Thu, 26 Mar 2015 22:05:20 +0000 (23:05 +0100)]
software/bios/sdram: make seed_to_data static
Florent Kermarrec [Thu, 26 Mar 2015 22:02:23 +0000 (23:02 +0100)]
sdram/phy/simphy: remove use of iter
Florent Kermarrec [Thu, 26 Mar 2015 21:24:47 +0000 (22:24 +0100)]
sdram/phy: add simphy (software memtest OK in simulation with MT48LC4M16)
Florent Kermarrec [Thu, 26 Mar 2015 21:16:31 +0000 (22:16 +0100)]
software/bios/sdram: select the type of data we want to generate for memtest with TEST_RANDOM_DATA (debugging hardware is easier with a simple counter)
Florent Kermarrec [Wed, 25 Mar 2015 22:59:29 +0000 (23:59 +0100)]
software/memtest: remove Mixxeo/M1 hardcoded values in bandwidth computation
Florent Kermarrec [Wed, 25 Mar 2015 18:00:07 +0000 (19:00 +0100)]
sofware/memtest: use MAIN_RAM_SIZE from mem.h
Florent Kermarrec [Wed, 25 Mar 2015 17:44:08 +0000 (18:44 +0100)]
tools/flterm.py: small clean up
Florent Kermarrec [Wed, 25 Mar 2015 16:57:42 +0000 (17:57 +0100)]
libcompiler-rt: add ucmpdi2.o
Florent Kermarrec [Wed, 25 Mar 2015 16:25:20 +0000 (17:25 +0100)]
sofware/memtest: update bandwidth registers
Florent Kermarrec [Wed, 25 Mar 2015 16:22:26 +0000 (17:22 +0100)]
sdram/core/lasmicon: automatically insert bandwidth module when with_memtest is True
Florent Kermarrec [Tue, 24 Mar 2015 17:26:18 +0000 (18:26 +0100)]
sdram: pass module as phy parameter, define memtype in modules and only keep phy parameter in register_sdram_phy
Florent Kermarrec [Tue, 24 Mar 2015 16:25:59 +0000 (17:25 +0100)]
sdram: use names that are more explicit for bank_a, row_a,...: bankbits, rowbits, .... Add databits to GeomSettings.
Florent Kermarrec [Wed, 25 Mar 2015 09:59:31 +0000 (10:59 +0100)]
tools: add minimal flterm.py (basic flterm.c clone with kernel loading for now)
flterm.c is not portable, we need a portable alternative. Once flterm.py will support all flterm features, it will be possible to remove flterm.c.
Florent Kermarrec [Wed, 25 Mar 2015 15:39:30 +0000 (16:39 +0100)]
linker-sdram.ld: sdram mem region is now called main_ram
Florent Kermarrec [Sun, 22 Mar 2015 10:08:47 +0000 (11:08 +0100)]
liteusb: give more generic names to modules: FtdiXXX becomes LiteUSBXXX, move PHY outside of core (builds on minispartan6)
Florent Kermarrec [Sun, 22 Mar 2015 09:56:56 +0000 (10:56 +0100)]
liteusb: make oe_n optional on ft2232h phy
Florent Kermarrec [Sun, 22 Mar 2015 09:56:29 +0000 (10:56 +0100)]
liteusb: fix imports
Florent Kermarrec [Sun, 22 Mar 2015 02:29:11 +0000 (03:29 +0100)]
targets: add minispartan6 (SDRAM working)
Florent Kermarrec [Sun, 22 Mar 2015 02:20:02 +0000 (03:20 +0100)]
sdram/module: fix tREFI on AS4C16M16
Florent Kermarrec [Sun, 22 Mar 2015 07:32:38 +0000 (08:32 +0100)]
targets: pipistrello/ppro, fix stupid mistake 10ex --> 1ex...
Florent Kermarrec [Sat, 21 Mar 2015 23:30:21 +0000 (00:30 +0100)]
targets: fix CLKIN1_PERIOD on ppro and pipistrello
Florent Kermarrec [Sat, 21 Mar 2015 21:51:24 +0000 (22:51 +0100)]
sdram: pass sdram_controller_settings to SDRAMSoC
Florent Kermarrec [Sat, 21 Mar 2015 20:32:39 +0000 (21:32 +0100)]
sdram: simplify the way we pass settings to controller and rename ramcon_type to sdram_controller_type (more explicit)
Florent Kermarrec [Sat, 21 Mar 2015 20:00:12 +0000 (21:00 +0100)]
rename sdram mapping to main_ram