Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
W
White Rabbit core collection
Manage
Activity
Members
Labels
Plan
Issues
33
Issue boards
Milestones
Wiki
Code
Merge requests
3
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Deploy
Releases
Monitor
Incidents
Service Desk
Analyze
Value stream analytics
Contributor analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Projects
White Rabbit core collection
Commits
f086ba41
Commit
f086ba41
authored
11 years ago
by
Grzegorz Daniluk
Browse files
Options
Downloads
Patches
Plain Diff
switch-optimization: rx_bypass_queue can use generic fifo or xilinx fifo
parent
6d5258db
Branches
hl-wrpc-v5.0
No related merge requests found
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
modules/wr_endpoint/ep_rx_bypass_queue.vhd
+91
-38
91 additions, 38 deletions
modules/wr_endpoint/ep_rx_bypass_queue.vhd
modules/wr_endpoint/ep_rx_crc_size_check.vhd
+4
-2
4 additions, 2 deletions
modules/wr_endpoint/ep_rx_crc_size_check.vhd
with
95 additions
and
40 deletions
modules/wr_endpoint/ep_rx_bypass_queue.vhd
+
91
−
38
View file @
f086ba41
...
...
@@ -36,11 +36,13 @@
library
ieee
;
use
ieee
.
std_logic_1164
.
all
;
use
work
.
genram_pkg
.
all
;
entity
ep_rx_bypass_queue
is
generic
(
g_size
:
integer
:
=
3
;
g_width
:
integer
:
=
18
);
g_width
:
integer
:
=
18
;
g_xilinx_fifo
:
boolean
:
=
false
);
port
(
rst_n_i
:
in
std_logic
;
...
...
@@ -90,7 +92,7 @@ architecture behavioral of ep_rx_bypass_queue is
signal
q_valid
:
std_logic_vector
(
g_size
-1
downto
0
);
signal
qempty
,
qfull
:
std_logic
;
signal
qempty
,
qfull
,
q_almost_full
:
std_logic
;
signal
flushing
:
std_logic
;
signal
valid_mask
:
std_logic
;
signal
valid_int
:
std_logic
;
...
...
@@ -99,54 +101,105 @@ architecture behavioral of ep_rx_bypass_queue is
begin
-- behavioral
qempty
<=
f_queue_occupation
(
q_valid
,
'1'
);
qfull
<=
f_queue_occupation
(
q_valid
,
'0'
);
GEN_ORIG
:
if
g_xilinx_fifo
=
false
generate
qempty
<=
f_queue_occupation
(
q_valid
,
'1'
);
qfull
<=
f_queue_occupation
(
q_valid
,
'0'
);
empty_o
<=
qempty
;
empty_o
<=
qempty
;
gen_sreg
:
for
i
in
0
to
g_width
-1
generate
gen_sreg
:
for
i
in
0
to
g_width
-1
generate
U_sreg
:
ep_shift_reg
generic
map
(
g_size
=>
g_size
)
port
map
(
clk_i
=>
clk_i
,
ce_i
=>
sreg_enable
,
d_i
=>
d_i
(
i
),
q_o
=>
q_o
(
i
));
U_sreg
:
ep_shift_reg
generic
map
(
g_size
=>
g_size
)
port
map
(
clk_i
=>
clk_i
,
ce_i
=>
sreg_enable
,
d_i
=>
d_i
(
i
),
q_o
=>
q_o
(
i
));
end
generate
gen_sreg
;
end
generate
gen_sreg
;
sreg_enable
<=
'1'
when
((
valid_i
=
'1'
)
or
(
qempty
=
'0'
and
(
flushing
=
'1'
)
and
valid_int
=
'1'
))
else
'0'
;
sreg_enable
<=
'1'
when
((
valid_i
=
'1'
)
or
(
qempty
=
'0'
and
(
flushing
=
'1'
)
and
valid_int
=
'1'
))
else
'0'
;
p_queue
:
process
(
clk_i
)
begin
if
rising_edge
(
clk_i
)
then
if
rst_n_i
=
'0'
or
purge_i
=
'1'
then
flushing
<=
'0'
;
valid_mask
<=
'0'
;
q_valid
<=
(
others
=>
'0'
);
else
if
(
flushing
=
'1'
and
qempty
=
'1'
)
then
flushing
<=
'0'
;
elsif
(
flush_i
=
'1'
and
qempty
=
'0'
)
then
flushing
<=
'1'
;
end
if
;
p_queue
:
process
(
clk_i
)
begin
if
rising_edge
(
clk_i
)
then
if
rst_n_i
=
'0'
or
purge_i
=
'1'
then
flushing
<=
'0'
;
valid_mask
<=
'0'
;
q_valid
<=
(
others
=>
'0'
);
else
if
(
flushing
=
'1'
and
qempty
=
'1'
)
then
flushing
<=
'0'
;
elsif
(
flush_i
=
'1'
and
qempty
=
'0'
)
then
flushing
<=
'1'
;
end
if
;
valid_mask
<=
dreq_i
;
valid_mask
<=
dreq_i
;
if
sreg_enable
=
'1'
then
q_valid
(
0
)
<=
valid_i
;
q_valid
(
q_valid
'length
-1
downto
1
)
<=
q_valid
(
q_valid
'length
-2
downto
0
);
if
sreg_enable
=
'1'
then
q_valid
(
0
)
<=
valid_i
;
q_valid
(
q_valid
'length
-1
downto
1
)
<=
q_valid
(
q_valid
'length
-2
downto
0
);
end
if
;
end
if
;
end
if
;
end
if
;
end
process
;
end
process
;
dreq_o
<=
dreq_i
and
not
(
flush_i
or
flushing
);
valid_int
<=
(
qfull
and
valid_i
)
or
(
not
qempty
and
flushing
and
valid_mask
);
valid_o
<=
valid_int
;
end
generate
;
GEN_XIL
:
if
g_xilinx_fifo
=
true
generate
U_FIFO
:
generic_sync_fifo
generic
map
(
g_data_width
=>
g_width
,
g_size
=>
g_size
+
1
,
g_show_ahead
=>
true
,
g_with_empty
=>
true
,
g_with_full
=>
true
,
g_with_almost_empty
=>
false
,
g_with_almost_full
=>
true
,
g_with_count
=>
false
,
g_almost_full_threshold
=>
3
)
port
map
(
rst_n_i
=>
rst_n_i
,
clk_i
=>
clk_i
,
d_i
=>
d_i
,
we_i
=>
valid_i
,
q_o
=>
q_o
,
rd_i
=>
valid_int
,
--read when fifo is full or when flushing requested
empty_o
=>
qempty
,
full_o
=>
open
,
almost_full_o
=>
qfull
);
dreq_o
<=
dreq_i
and
not
(
flush_i
or
flushing
);
valid_int
<=
(
qfull
and
valid_i
)
or
(
not
qempty
and
flushing
and
valid_mask
);
valid_o
<=
valid_int
;
empty_o
<=
qempty
;
--flushing process
process
(
clk_i
)
begin
if
rising_edge
(
clk_i
)
then
if
rst_n_i
=
'0'
or
purge_i
=
'1'
then
flushing
<=
'0'
;
valid_mask
<=
'0'
;
else
if
(
flushing
=
'1'
and
qempty
=
'1'
)
then
flushing
<=
'0'
;
elsif
(
flush_i
=
'1'
and
qempty
=
'0'
)
then
flushing
<=
'1'
;
end
if
;
valid_mask
<=
dreq_i
;
end
if
;
end
if
;
end
process
;
dreq_o
<=
dreq_i
and
not
(
flush_i
or
flushing
);
valid_int
<=
(
qfull
and
valid_i
)
or
(
not
qempty
and
flushing
and
valid_mask
);
valid_o
<=
valid_int
;
end
generate
;
end
behavioral
;
...
...
This diff is collapsed.
Click to expand it.
modules/wr_endpoint/ep_rx_crc_size_check.vhd
+
4
−
2
View file @
f086ba41
...
...
@@ -41,7 +41,8 @@ architecture behavioral of ep_rx_crc_size_check is
component
ep_rx_bypass_queue
generic
(
g_size
:
integer
;
g_width
:
integer
);
g_width
:
integer
;
g_xilinx_fifo
:
boolean
:
=
false
);
port
(
rst_n_i
:
in
std_logic
;
clk_i
:
in
std_logic
;
...
...
@@ -139,7 +140,8 @@ begin -- behavioral
U_bypass_queue
:
ep_rx_bypass_queue
generic
map
(
g_size
=>
3
,
g_width
=>
18
)
g_width
=>
18
,
g_xilinx_fifo
=>
false
)
port
map
(
rst_n_i
=>
rst_n_i
,
clk_i
=>
clk_sys_i
,
...
...
This diff is collapsed.
Click to expand it.
Preview
0%
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment