Commit Graph

227 Commits (6bc1eae48d933701833b93fb811fcea07607ecd2)

Author SHA1 Message Date
Lars-Peter Clausen 6bc1eae48d axi_dmac: 2d_transfer: Remove resets from data path
The data path register of the 2d_transfer module are qualified by the
corresponding valid signal. Their content is not used until they have been
explicitly initialized. There is no need to reset them explicitly.

This reduces the fan-out of the reset signal.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-06-05 14:28:40 +02:00
Lars-Peter Clausen 6b7a46410c axi_dmac: address_generator: Remove resets from data path
There is no need to reset the data path in the address generator. The
values of the register on the data path are not used until they have been
explicitly initialized. Removing the reset simplifies the structure and
reduces the fan-out of the reset signal.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-06-05 14:28:40 +02:00
Lars-Peter Clausen 67600f9831 axi_dmac: Use localparam instead of parameter
Xilinx tools don't allow to use $clog2() when computing the value of a
localparam, even though it is valid Verilog.

For this reason a parameter was used for BYTES_PER_BURST_WIDTH so far. But
that generates warnings from both Quartus and Vivado since the parameter is
not part of the parameter list.

Fix this by changing it to a localparam and computing the log2() manually.
The upper limit for the burst length is known to be 4k, so values larger
than that don't have to be supported.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-06-05 14:28:40 +02:00
Lars-Peter Clausen cf5208122a axi_dmac: Increase default store-and-forward memory size to 8 bursts
A larger store-and-forward memory provides better protection against worst
case memory interface latencies by being able to store more data before
over-/underflowing.

Based on empirical testing it was found that using a size of 4 bursts can
still result in underflows/overflows under certain conditions. These do not
happen when using a size of 8 bursts.

This change does not significantly increase resource consumption. Both on
Intel and Xilinx the block RAM has a minimum depth of 512 entries. With a
default burst length of 16 beats that allows for up to 32 bursts without
requiring additional block RAM.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-06-05 14:28:40 +02:00
Lars-Peter Clausen b18b16fd3a axi_dmac: Use a more descriptive label for the store-and-forward memory size
The label for the store-and-forward memory size configuration option at the
moment is just "FIFO Size" and while the store-and-forward memory uses a
FIFO that is just a implementation detail.

Change the label to "Store-and-Forward Memory Size". This is more
descriptive as it references the function not the implementation.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-06-05 14:28:40 +02:00
Lars-Peter Clausen 15b0e38e23 axi_dmac: List valid store-and-forward memory sizes
For correct operation the store-and-forward memory size must be a
power-of-two in the range of 2 to 32.

This is simple enough so we can list all values and let the IP Integrator
and QSYS perform proper validation of the parameter.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-06-05 14:28:40 +02:00
Lars-Peter Clausen 682895c7ae axi_dmac: dest_axi_stream: Remove outdated comment
This comment hasn't been true in a long long time. It does not have any
relation to the code around it anymore.

So just remove it.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-06-05 14:28:40 +02:00
Lars-Peter Clausen 7a804c1d76 axi_dmac: Fix debug ID order
Commit e6aacd2f56 ("axi_dmac: Better support debug IDs when ID_WIDTH !=
3") managed to get the order of the IDs in the debug register wrong.
Restore the original order.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-05-04 12:40:15 +02:00
Laszlo Nagy 3f94fec620 axi_dmac/dma_write_tb: added data integrity check
Added passing condition checking.
Added variable request sizes
2018-05-03 14:49:06 +02:00
Laszlo Nagy 5c2e10eca5 axi_dmac: added ModelSim support to run_tb.sh
Modified run_tb.sh to work with Modelsim if the SIMULATOR variable is set to
"modelsim"
2018-05-03 14:49:06 +02:00
Laszlo Nagy ee4932ecee axi_dmac: made vlog pass
Did minor changes to the code to make ModelSim vlog pass.
2018-05-03 14:49:06 +02:00
Lars-Peter Clausen 24d17e8bcc axi_dmac: Add transfer testbenches
Add simple transfer testbenches that test the read and write to AXI memory
paths of the DMAC.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-05-03 14:49:06 +02:00
Lars-Peter Clausen b3f027fc89 axi_dmac: Add simple register map testbench
Add a testbench that exercises the basic functionality of the axi_dmac
register map module.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-05-03 14:49:06 +02:00
Lars-Peter Clausen ef3687e99d axi_dmac: Split register map into separate sub-module
Split the register map code into a separate sub-module instead of having it
as part of the top-level axi_dmac.v file.

This makes it easier to component test the register map behavior
independently from the DMA transfer logic.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-05-03 14:49:06 +02:00
Lars-Peter Clausen ccb69e71a3 axi_dmac: axi_dmac_hw.tcl: Use ad_ip_files helper
Use the ad_ip_files helper to reduce the amount of boiler plate code.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-05-03 14:49:06 +02:00
Lars-Peter Clausen ee57f869f1 axi_dmac: Fix bus resize block reset
When the source and destination bus widths don't match a resize block is
inserted on the side of the narrower bus. This resize block can contain
partial data.

To ensure that there is no residual partial data is left in the resize
block after a transfer shutdown the resize block is reset when the DMA is
disabled.

Currently this is implemented by tying the reset signal of the resize block
to the enable signal of the DMA. This enable signal is only a indicator
though that the DMA should shutdown. For a proper shutdown outstanding
transactions still need to be completed.

The data that is in the resize block might be required to complete those
transactions. So performing the reset when the enable signal goes low can
lead to a situation where the DMA tries to complete a transaction but can't
do it because the data required to do so has been erased by resetting the
resize block. This leads to a dead lock and the system has to be rebooted
to recover from it.

To solve this use the sync_id signal to reset the resize block. The sync_id
signal will only be asserted when both the destination and source side
module have indicated that they are ready to be reset and there are no more
pending transactions.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-05-02 09:37:02 +02:00
Lars-Peter Clausen f3102eea5a axi_dmac: Limit MAX_BYTES_PER_BURST to maximum supported value
The MAX_BYTES_PER_BURST option allows to configure the maximum bytes that
are part of a burst. This can be an arbitrary value.

At the same time there is a limit of how many bytes can be supported by the
memory buses. A AXI3 interface supports a maximum of 16 beats per burst
and a AXI4 interface supports a maximum of 256 beats per burst.

At the moment the it is possible to specify a MAX_BYTES_PER_BURST value
that exceeds what can be supported by the AXI memory-mapped bus. If that is
the case undefined behavior will occur and the DMAC will function
incorrectly.

To avoid this make sure that the MAX_BYTES_PER_BURST value does not exceed
the maximum that can be supported by the interfaces.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-24 12:49:24 +02:00
Lars-Peter Clausen 5418dfeb50 axi_dmac: axi_dmac_hw.tcl: Fix indention
Fix some slight indentation issues in the axi_dmac_hw.tcl.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-24 12:49:24 +02:00
Lars-Peter Clausen e609c7fd3b axi_dmac: Prevent destination AXI burst length truncation
The width of the AXI burst length field depends on the AXI standard
version. For AXI3 the width is 4 bits allowing a maximum burst length of 16
beats, for AXI4 it is 8 bits wide allowing a maximum burst length of 256
beats.

At the moment the width of the length signals are determined by type of the
source AXI interface, even if the source interface type is not AXI. This
means if the source interface is set to AXI3 and the destination interface
is set to AXI4 the internal width of the signal for all interfaces will be
4 bits. This leads to a truncation of the destination bus length field,
which is supposed to be 8 bits.

If burst are generated that are longer than 16 beats the upper bits of the
length signal will be truncated. The result of this will be that the
external AXI slave interface (e.g. the DDR memory) and the internal logic
in the DMA disagree about burst length.  The DMA will eventually lock up
when its internal buffers are full.

To avoid this issue have different configuration parameters for the source
and destination interface that configure the AXI bus length field width.

This way one of the interfaces can be configured for AXI3 and the other for
AXI4 without interfering with each other.

Fixes: commit 495d2f3056 ("axi_dmac: Propagate awlen/arlen width through the core")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-24 12:49:24 +02:00
Laszlo Nagy bfac56199e axi_dmac: adding missing dependency for Intel flow
When the DMAC is used in async clock domains the data FIFO instantiate
an ad_mem component to handle properly the clock crossing.
For Intel, this mode is used only in FMCJESDADC1 designs but without this
an error could appear in other projects too if the user reconfigures the core.
2018-04-17 16:34:41 +03:00
Laszlo Nagy c42ed7dd52 axi_dmac: removed harmful SDC constraint
The set_false_path constraint targeted to the *ram* cells of the dmac
matched several intra clock domain paths where the timing analysis got
ignored resulting in intermitent data integrity issues.
2018-04-17 16:34:41 +03:00
Laszlo Nagy ad05a5ecc1 axi_dmac: AXI3 support on Intel qsys
Exposed AXI3 interface on the Intel version of the IP for UI and feature consistency.
Some of the signals that are defined as optional in the AMBA standard
are marked as mandatory in Qsys in case of AXI3. Because of this such signals
were added to the interface of the DMAC and driven with default values.

For Xilinx in order to keep existing behavior the newly added signals
are hidden from the interface.

New parameters are added to define the width of the AXI transaction IDs;
these are hidden from the UI; We can add them to the UI if the fixed size
of the IDs will cause port incompatibility issues.
2018-04-17 15:12:01 +03:00
Lars-Peter Clausen 2c4d81a221 axi_dmac: Disable 2D transfer support by default
The primary use-case of the DMA controller is in non-2D mode. Make this the
default, since allows projects to instantiate the controller with the
default configuration without having to explicitly disable 2D support.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-11 15:09:54 +03:00
Lars-Peter Clausen 05e8604ea7 axi_dmac: Remove unused pause signal from address generator
The pause signal is not used inside the address generator module. Remove
it.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-11 15:09:54 +03:00
Lars-Peter Clausen 2b2c1f6a1e axi_dmac: Fix some indentation errors
Purely cosmetic.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-11 15:09:54 +03:00
Lars-Peter Clausen 2b914d33c1 Move Altera IP core dependency tracking to library Makefiles
Currently the individual IP core dependencies are tracked inside the
library Makefile for Xilinx IPs and the project Makefiles only reference
the IP cores.

For Altera on the other hand the individual dependencies are tracked inside
the project Makefile. This leads to a lot of duplicated lists and also
means that the project Makefiles need to be regenerated when one of the IP
cores changes their files.

Change the Altera projects to a similar scheme than the Xilinx projects.
The projects themselves only reference the library as a whole as their
dependency while the library Makefile references the individual source
dependencies.

Since on Altera there is no target that has to be generated create a dummy
target called ".timestamp_altera" who's only purpose is to have a timestamp
that is greater or equal to the timestamp of all of the IP core files. This
means the project Makefile can have a dependency on this file and make sure
that the project will be rebuild if any of the files in the library
changes.

This patch contains quite a bit of churn, but hopefully it reduces the
amount of churn in the future when modifying Altera IP cores.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-11 15:09:54 +03:00
Lars-Peter Clausen 48ef19ec60 library: Track additional file types as dependency in Makefile
Re-generate the Makefile with some additional file types added as
dependencies to the IP cores.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-11 15:09:54 +03:00
Lars-Peter Clausen 3dac544d0a axi_dmac_ip.tcl: Add include files to file list
The include files are currently only implicitly added to the component file
list. Do it explicitly as this will make sure that they show up in the
generated Makefile dependency list.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-11 15:09:54 +03:00
Lars-Peter Clausen 35a39ba2e6 Regenerate library Makefiles using the new shared Makefile include
This reduces the amount of boilerplate code that is present in these
Makefiles by a lot.

It also makes it possible to update the Makefile rules in future without
having to re-generate all the Makefiles.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-04-11 15:09:54 +03:00
Istvan Csomortani 425e803364 license: Fix a spelling mistake 2018-04-11 15:09:54 +03:00
Laszlo Nagy 5cba46165a axi_dmac: fix synthesis warnings
Separated the 2D transfer registers to a separate generate block
2018-04-11 15:09:54 +03:00
Adrian Costina 5bfc585524 axi_dmac: Added MAX_BYTES_PER_BURST and DISABLE_DEBUG_REGISTERS parameters to Intel IP 2018-04-11 15:09:54 +03:00
Istvan Csomortani d13ff8df1e axi_dmac: In SDP mode REGCEB is connected to GND
In newer version of Vivado (e.g. 2017.4) the REGCEB pin of the block ram
macro is connected to ground. So the following false path became
redundant.
2018-04-11 15:09:54 +03:00
Istvan Csomortani a740b6012f Make: Use $(MAKE) for recursive make commands
This commit should resolve the issue #64.

Recursive make commands should always use the variable MAKE, not the explicit
command name ‘make’.
2018-03-07 07:40:19 +00:00
Matt Fornero 3e7399913f axi_dmac: Include TLAST in AXIS slave port
Bundle the TLAST signal in with the other AXIS slave signals to enable
easier connection between AXIS devices that use TLAST

Signed-off-by: Matt Fornero <matt.fornero@mathworks.com>
2018-01-23 17:43:48 +01:00
Lars-Peter Clausen c6073954d9 axi_dmac: Add limited TLAST support for streaming AXI source interface
Add some limit TLAST support for the streaming AXI source interface. An
asserted TLAST signal marks the end of a packet and the following data beat
is the first beat for the next packet.

Currently the DMAC does not support for completing a transfer before all
requested bytes have been transferred. So the way this limited TLAST
support is implemented is by filling the remainder of the buffer with 0x00.

While the DMAC is busy filling the buffer with zeros back-pressure is
asserted on the external streaming AXI interface by keeping TREADY
de-asserted.

The end of a buffer is marked by a transfer that has the last bit set in
the FLAGS control register.

In the future we might add support for transfer completion before all
requested bytes have been transferred.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2018-01-23 17:43:48 +01:00
Istvan Csomortani 2b2c6b57f9 axi_dmac: Align the data_ready to data
The commit 6900c have added an additional register stage into the fifo read
data path, but the control signals (ready/valid/underflow) were not realigned
to the data. This can cause data lose or duplicated samples in some case.
Realign the control signals to the data.
2017-11-21 13:15:03 +00:00
Istvan Csomortani 6900c9979b axi_dmac: Reset fifo_rd_data without delaying the valid data 2017-11-03 13:41:50 +00:00
Istvan Csomortani 06bab87733 axi_dmac: Reset fifo_rd_data when DMA is off - v2
The first attempt (f3daf0) faild miserably. When the data_req signal
from the device had more than 1 cycle of deassert state, because of the
added latency of the data stream, the device got 'zeros' too.
In this fix, the DMA will hold the valid data on the bus, between two
consecutive data request. The bus is reseted just after all the data
were sent out.
2017-10-10 08:10:24 +01:00
Istvan Csomortani f3daf0dacb axi_dmac: Reset the fifo_rd_data if the DMA is off
Reset the fifo_rd_data if the DMA does not have an active transfer.
Becasue all the DAC device cores are transfering the data from the FIFO
interface to the data interface without any validation signal, DMA needs to put
the data bus into a known state, to prevent the device core to send the
last known data again and again.
2017-10-05 08:54:15 +01:00
Istvan Csomortani 0064004d34 axi_dmac: Control s_axis_user/fifo_wr_sync validity
The ports s_axis_user or fifo_wr_sync will be active just
if the SYNC_TRANSFER_START is enabled.
2017-10-03 09:32:14 +01:00
Istvan Csomortani 08a31a7d9f axi_dmac: Fix the last incorrect merge 2017-10-03 09:15:45 +01:00
Istvan Csomortani 89bd8b44d4 Merge branch 'dev' into hdl_2017_r1 2017-09-26 07:42:19 +01:00
Lars-Peter Clausen e6aacd2f56 axi_dmac: Better support debug IDs when ID_WIDTH != 3
The current layout of the debug ID register assumes that the ID_WIDTH is 3.
Change things so that the padding 0 width depends on the ID_WIDTH
parameter so that we end up with the same register layout regardless of the
value of ID_WIDTH.

Also split things into two registers, this allows for an ID_WIDTH up to 8
(which should hopefully be enough for all practical applications).

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2017-08-24 17:42:20 +02:00
Lars-Peter Clausen d807b10632 axi_dmac: axi_dmac_hw.tcl: Set associated reset and addressable point for the interrupt interface
Altera recommends to set those properties for better qsys integration.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2017-08-13 10:28:11 +02:00
Lars-Peter Clausen 95d530e7c9 axi_dmac: Set axi4lite address space size to 4k
The AXI specification that the minimum address space size is 4k, make sure
the axi_dmac adheres to this.

Internally the register space is still 2k. This means the upper and lower
2k of the axi4lite register space will map to the same internal registers.
Software must not rely on this and only access the lower 2k to enable
compatibility in case the internal space grows in the future.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2017-08-01 15:22:29 +02:00
Lars-Peter Clausen 1bf25edf77 axi_dmac: src_axi_stream: Terminate data mover m_axi_last signal
Terminate the m_axi_list signal of the data mover instance in the
src_axi_stream module. This avoids a warning about the port being
unconnected.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2017-08-01 15:22:29 +02:00
Lars-Peter Clausen 699d8970dd axi_dmac: axi_dmac_hw.tcl: Disable unused interfaces instead of not creating them
Currently the axi_dmac_hw.tcl script does not create interfaces if they
are not used in the current configuration. This has the disadvantage that
the ports belonging to these interfaces are not included in the generated
HDL wrapper. Which will generate a fair bunch of warnings when synthesizing
the HDL.

Instead always generate all interfaces, but disable those that are not used
in the current configuration. This will make sure that the ports belonging
to these interfaces are properly tied-off in the generate wrapper HDL.

This reduces the amount of false positive warnings generated and makes it
easier to spot actual issues.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2017-08-01 15:22:29 +02:00
Lars-Peter Clausen 834eb6aaa5 axi_dmac: dest_axi_mm: Use fixed wstrb signal
The DMAC currently doesn't support transfers where the length is not a
multiple of the bus width. When generating the wstrb signal we do pretend
though that we do and dynamically generate it based on the LSBs of the
transfer length.

Given that the other parts of the DMA don't support such transfers this is
unnecessary though. So remove it for now and replace it with a constant
expression where wstrb is always fully asserted.

The generated logic for the wstrb signal was quite terrible, so this
improves the timing of the core.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2017-08-01 15:22:29 +02:00
Lars-Peter Clausen aeabe91144 axi_dmac: Comment out unused src_response interface
Currently the read side of the src_response interface is not used. This
leads to warnings about signals that have a value assigned but are never
read.

To avoid this just comment out all signals that are related to the
src_response interface for now.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2017-08-01 15:22:29 +02:00