github.com/antonblanchard.microwatt
1
0
Fork 0
mirror of https://github.com/antonblanchard/microwatt.git synced 2026-01-11 23:43:15 +00:00
Code Releases Activity
antonblanchard.microwatt/mmu.vhdl
Paul Mackerras f9e5622327 Move iTLB from icache to fetch1 
This moves the address translation step for instruction fetches one
cycle earlier, so that it now happens in the fetch1 stage.  There is
now a 2-entry mini translation cache ("ERAT", or effective to real
address translation cache) which operates on the output of the
multiplexer that selects the instruction address for the next cycle.
The ERAT consists of two effective address registers and two
corresponding real address registers.  They store the page number part
of the addresses for a 4kB page size, which is the smallest page size
supported by the architecture.

If the effective address doesn't match either of the EA registers, and
address translation is enabled, then i_out.req goes low for two cycles
while the iTLB is looked up.  Experimentally, this delay results in a
0.1% drop in coremark performance; allowing two cycles for the lookup
results in better timing.  The result from the iTLB is placed into the
least recently used ERAT entry and then used to translate the address
as normal.  If address translation is not enabled then the EA is used
directly as the real address.

The iTLB structure is the same as it was before; direct mapped,
indexed using a hashed EA.

The "fetch failed" signal, which indicates a TLB miss or protection
violation, is now generated in fetch1 and passed through icache.
When it is asserted, fetch1 goes into a stalled state until a PTE
arrives from the MMU (which gets put into both the iTLB and the ERAT),
or an interrupt or redirect occurs.

Any TLB invalidations from the MMU invalidate the whole ERAT.

Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
2023-09-19 17:34:47 +10:00

520 lines
18 KiB
VHDL
Raw Blame History

library ieee;
use ieee.std_logic_1164.all;
use ieee.numeric_std.all;
library work;
use work.common.all;
-- Radix MMU
-- Supports 4-level trees as in arch 3.0B, but not the two-step translation for
-- guests under a hypervisor (i.e. there is no gRA -> hRA translation).
entity mmu is
port (
clk : in std_ulogic;
rst : in std_ulogic;
l_in : in Loadstore1ToMmuType;
l_out : out MmuToLoadstore1Type;
d_out : out MmuToDcacheType;
d_in : in DcacheToMmuType;
i_out : out MmuToITLBType
);
end mmu;
architecture behave of mmu is
type state_t is (IDLE,
DO_TLBIE,
TLB_WAIT,
PART_TBL_READ,
PART_TBL_WAIT,
PART_TBL_DONE,
PROC_TBL_READ,
PROC_TBL_WAIT,
SEGMENT_CHECK,
RADIX_LOOKUP,
RADIX_READ_WAIT,
RADIX_LOAD_TLB,
RADIX_FINISH
);
type reg_stage_t is record
-- latched request from loadstore1
valid : std_ulogic;
iside : std_ulogic;
store : std_ulogic;
priv : std_ulogic;
addr : std_ulogic_vector(63 downto 0);
inval_all : std_ulogic;
-- config SPRs
ptcr : std_ulogic_vector(63 downto 0);
pid : std_ulogic_vector(31 downto 0);
-- internal state
state : state_t;
done : std_ulogic;
err : std_ulogic;
prtbl : std_ulogic_vector(63 downto 0);
ptb_valid : std_ulogic;
pgtbl0 : std_ulogic_vector(63 downto 0);
pt0_valid : std_ulogic;
pgtbl3 : std_ulogic_vector(63 downto 0);
pt3_valid : std_ulogic;
shift : unsigned(5 downto 0);
mask_size : unsigned(4 downto 0);
pgbase : std_ulogic_vector(55 downto 0);
pde : std_ulogic_vector(63 downto 0);
invalid : std_ulogic;
badtree : std_ulogic;
segerror : std_ulogic;
perm_err : std_ulogic;
rc_error : std_ulogic;
end record;
signal r, rin : reg_stage_t;
signal addrsh : std_ulogic_vector(15 downto 0);
signal mask : std_ulogic_vector(15 downto 0);
signal finalmask : std_ulogic_vector(43 downto 0);
begin
-- Multiplex internal SPR values back to loadstore1, selected
-- by l_in.sprnf.
l_out.sprval <= r.ptcr when l_in.sprnf = '1' else x"00000000" & r.pid;
mmu_0: process(clk)
begin
if rising_edge(clk) then
if rst = '1' then
r.state <= IDLE;
r.valid <= '0';
r.ptb_valid <= '0';
r.pt0_valid <= '0';
r.pt3_valid <= '0';
r.ptcr <= (others => '0');
r.pid <= (others => '0');
else
if rin.valid = '1' then
report "MMU got tlb miss for " & to_hstring(rin.addr);
end if;
if l_out.done = '1' then
report "MMU completing op without error";
end if;
if l_out.err = '1' then
report "MMU completing op with err invalid=" & std_ulogic'image(l_out.invalid) &
" badtree=" & std_ulogic'image(l_out.badtree);
end if;
if rin.state = RADIX_LOOKUP then
report "radix lookup shift=" & integer'image(to_integer(rin.shift)) &
" msize=" & integer'image(to_integer(rin.mask_size));
end if;
if r.state = RADIX_LOOKUP then
report "send load addr=" & to_hstring(d_out.addr) &
" addrsh=" & to_hstring(addrsh) & " mask=" & to_hstring(mask);
end if;
r <= rin;
end if;
end if;
end process;
-- Shift address bits 61--12 right by 0--47 bits and
-- supply the least significant 16 bits of the result.
addrshifter: process(all)
variable sh1 : std_ulogic_vector(30 downto 0);
variable sh2 : std_ulogic_vector(18 downto 0);
variable result : std_ulogic_vector(15 downto 0);
begin
case r.shift(5 downto 4) is
when "00" =>
sh1 := r.addr(42 downto 12);
when "01" =>
sh1 := r.addr(58 downto 28);
when others =>
sh1 := "0000000000000" & r.addr(61 downto 44);
end case;
case r.shift(3 downto 2) is
when "00" =>
sh2 := sh1(18 downto 0);
when "01" =>
sh2 := sh1(22 downto 4);
when "10" =>
sh2 := sh1(26 downto 8);
when others =>
sh2 := sh1(30 downto 12);
end case;
case r.shift(1 downto 0) is
when "00" =>
result := sh2(15 downto 0);
when "01" =>
result := sh2(16 downto 1);
when "10" =>
result := sh2(17 downto 2);
when others =>
result := sh2(18 downto 3);
end case;
addrsh <= result;
end process;
-- generate mask for extracting address fields for PTE address generation
addrmaskgen: process(all)
variable m : std_ulogic_vector(15 downto 0);
begin
-- mask_count has to be >= 5
m := x"001f";
if is_X(r.mask_size) then
m := (others => 'X');
else
for i in 5 to 15 loop
if i < to_integer(r.mask_size) then
m(i) := '1';
end if;
end loop;
end if;
mask <= m;
end process;
-- generate mask for extracting address bits to go in TLB entry
-- in order to support pages > 4kB
finalmaskgen: process(all)
variable m : std_ulogic_vector(43 downto 0);
begin
m := (others => '0');
for i in 0 to 43 loop
if is_X(r.shift) then
m(i) := 'X';
elsif i < to_integer(r.shift) then
m(i) := '1';
end if;
end loop;
finalmask <= m;
end process;
mmu_1: process(all)
variable v : reg_stage_t;
variable dcreq : std_ulogic;
variable tlb_load : std_ulogic;
variable itlb_load : std_ulogic;
variable tlbie_req : std_ulogic;
variable ptbl_rd : std_ulogic;
variable prtbl_rd : std_ulogic;
variable pt_valid : std_ulogic;
variable effpid : std_ulogic_vector(31 downto 0);
variable prtable_addr : std_ulogic_vector(63 downto 0);
variable rts : unsigned(5 downto 0);
variable mbits : unsigned(5 downto 0);
variable pgtable_addr : std_ulogic_vector(63 downto 0);
variable pte : std_ulogic_vector(63 downto 0);
variable tlb_data : std_ulogic_vector(63 downto 0);
variable nonzero : std_ulogic;
variable pgtbl : std_ulogic_vector(63 downto 0);
variable perm_ok : std_ulogic;
variable rc_ok : std_ulogic;
variable addr : std_ulogic_vector(63 downto 0);
variable data : std_ulogic_vector(63 downto 0);
begin
v := r;
v.valid := '0';
dcreq := '0';
v.done := '0';
v.err := '0';
v.invalid := '0';
v.badtree := '0';
v.segerror := '0';
v.perm_err := '0';
v.rc_error := '0';
tlb_load := '0';
itlb_load := '0';
tlbie_req := '0';
v.inval_all := '0';
ptbl_rd := '0';
prtbl_rd := '0';
-- Radix tree data structures in memory are big-endian,
-- so we need to byte-swap them
for i in 0 to 7 loop
data(i * 8 + 7 downto i * 8) := d_in.data((7 - i) * 8 + 7 downto (7 - i) * 8);
end loop;
case r.state is
when IDLE =>
if l_in.addr(63) = '0' then
pgtbl := r.pgtbl0;
pt_valid := r.pt0_valid;
else
pgtbl := r.pgtbl3;
pt_valid := r.pt3_valid;
end if;
-- rts == radix tree size, # address bits being translated
rts := unsigned('0' & pgtbl(62 downto 61) & pgtbl(7 downto 5));
-- mbits == # address bits to index top level of tree
mbits := unsigned('0' & pgtbl(4 downto 0));
-- set v.shift to rts so that we can use finalmask for the segment check
v.shift := rts;
v.mask_size := mbits(4 downto 0);
v.pgbase := pgtbl(55 downto 8) & x"00";
if l_in.valid = '1' then
v.addr := l_in.addr;
v.iside := l_in.iside;
v.store := not (l_in.load or l_in.iside);
v.priv := l_in.priv;
if l_in.tlbie = '1' then
-- Invalidate all iTLB/dTLB entries for tlbie with
-- RB[IS] != 0 or RB[AP] != 0, or for slbia
v.inval_all := l_in.slbia or l_in.addr(11) or l_in.addr(10) or
l_in.addr(7) or l_in.addr(6) or l_in.addr(5);
-- RIC=2 or 3 flushes process table caches.
if l_in.ric(1) = '1' then
v.pt0_valid := '0';
v.pt3_valid := '0';
v.ptb_valid := '0';
end if;
v.state := DO_TLBIE;
else
v.valid := '1';
if r.ptb_valid = '0' then
-- need to fetch process table base from partition table
v.state := PART_TBL_READ;
elsif pt_valid = '0' then
-- need to fetch process table entry
-- set v.shift so we can use finalmask for generating
-- the process table entry address
v.shift := unsigned('0' & r.prtbl(4 downto 0));
v.state := PROC_TBL_READ;
elsif mbits = 0 then
-- Use RPDS = 0 to disable radix tree walks
v.state := RADIX_FINISH;
v.invalid := '1';
else
v.state := SEGMENT_CHECK;
end if;
end if;
end if;
if l_in.mtspr = '1' then
-- Move to PID needs to invalidate L1 TLBs and cached
-- pgtbl0 value. Move to PTCR does that plus
-- invalidating the cached pgtbl3 and prtbl values as well.
if l_in.sprnt = '0' then
v.pid := l_in.rs(31 downto 0);
else
v.ptcr := l_in.rs;
v.pt3_valid := '0';
v.ptb_valid := '0';
end if;
v.pt0_valid := '0';
v.inval_all := '1';
v.state := DO_TLBIE;
end if;
when DO_TLBIE =>
dcreq := '1';
tlbie_req := '1';
v.state := TLB_WAIT;
when TLB_WAIT =>
if d_in.done = '1' then
v.state := RADIX_FINISH;
end if;
when PART_TBL_READ =>
dcreq := '1';
ptbl_rd := '1';
v.state := PART_TBL_WAIT;
when PART_TBL_WAIT =>
if d_in.done = '1' then
v.prtbl := data;
v.ptb_valid := '1';
v.state := PART_TBL_DONE;
end if;
when PART_TBL_DONE =>
v.shift := unsigned('0' & r.prtbl(4 downto 0));
v.state := PROC_TBL_READ;
when PROC_TBL_READ =>
dcreq := '1';
prtbl_rd := '1';
v.state := PROC_TBL_WAIT;
when PROC_TBL_WAIT =>
if d_in.done = '1' then
if r.addr(63) = '1' then
v.pgtbl3 := data;
v.pt3_valid := '1';
else
v.pgtbl0 := data;
v.pt0_valid := '1';
end if;
-- rts == radix tree size, # address bits being translated
rts := unsigned('0' & data(62 downto 61) & data(7 downto 5));
-- mbits == # address bits to index top level of tree
mbits := unsigned('0' & data(4 downto 0));
-- set v.shift to rts so that we can use finalmask for the segment check
v.shift := rts;
v.mask_size := mbits(4 downto 0);
v.pgbase := data(55 downto 8) & x"00";
if mbits = 0 then
v.state := RADIX_FINISH;
v.invalid := '1';
else
v.state := SEGMENT_CHECK;
end if;
end if;
if d_in.err = '1' then
v.state := RADIX_FINISH;
v.badtree := '1';
end if;
when SEGMENT_CHECK =>
mbits := '0' & r.mask_size;
v.shift := r.shift + (31 - 12) - mbits;
nonzero := or(r.addr(61 downto 31) and not finalmask(30 downto 0));
if r.addr(63) /= r.addr(62) or nonzero = '1' then
v.state := RADIX_FINISH;
v.segerror := '1';
elsif mbits < 5 or mbits > 16 or mbits > (r.shift + (31 - 12)) then
v.state := RADIX_FINISH;
v.badtree := '1';
else
v.state := RADIX_LOOKUP;
end if;
when RADIX_LOOKUP =>
dcreq := '1';
v.state := RADIX_READ_WAIT;
when RADIX_READ_WAIT =>
if d_in.done = '1' then
v.pde := data;
-- test valid bit
if data(63) = '1' then
-- test leaf bit
if data(62) = '1' then
-- check permissions and RC bits
perm_ok := '0';
if r.priv = '1' or data(3) = '0' then
if r.iside = '0' then
perm_ok := data(1) or (data(2) and not r.store);
else
-- no IAMR, so no KUEP support for now
-- deny execute permission if cache inhibited
perm_ok := data(0) and not data(5);
end if;
end if;
rc_ok := data(8) and (data(7) or not r.store);
if perm_ok = '1' and rc_ok = '1' then
v.state := RADIX_LOAD_TLB;
else
v.state := RADIX_FINISH;
v.perm_err := not perm_ok;
-- permission error takes precedence over RC error
v.rc_error := perm_ok;
end if;
else
mbits := unsigned('0' & data(4 downto 0));
if mbits < 5 or mbits > 16 or mbits > r.shift then
v.state := RADIX_FINISH;
v.badtree := '1';
else
v.shift := v.shift - mbits;
v.mask_size := mbits(4 downto 0);
v.pgbase := data(55 downto 8) & x"00";
v.state := RADIX_LOOKUP;
end if;
end if;
else
-- non-present PTE, generate a DSI
v.state := RADIX_FINISH;
v.invalid := '1';
end if;
end if;
if d_in.err = '1' then
v.state := RADIX_FINISH;
v.badtree := '1';
end if;
when RADIX_LOAD_TLB =>
tlb_load := '1';
if r.iside = '0' then
dcreq := '1';
v.state := TLB_WAIT;
else
itlb_load := '1';
v.state := IDLE;
end if;
when RADIX_FINISH =>
v.state := IDLE;
end case;
if v.state = RADIX_FINISH or (v.state = RADIX_LOAD_TLB and r.iside = '1') then
v.err := v.invalid or v.badtree or v.segerror or v.perm_err or v.rc_error;
v.done := not v.err;
end if;
if r.addr(63) = '1' then
effpid := x"00000000";
else
effpid := r.pid;
end if;
prtable_addr := x"00" & r.prtbl(55 downto 36) &
((r.prtbl(35 downto 12) and not finalmask(23 downto 0)) or
(effpid(31 downto 8) and finalmask(23 downto 0))) &
effpid(7 downto 0) & "0000";
pgtable_addr := x"00" & r.pgbase(55 downto 19) &
((r.pgbase(18 downto 3) and not mask) or (addrsh and mask)) &
"000";
pte := x"00" &
((r.pde(55 downto 12) and not finalmask) or (r.addr(55 downto 12) and finalmask))
& r.pde(11 downto 0);
-- update registers
rin <= v;
-- drive outputs
if tlbie_req = '1' then
addr := r.addr;
tlb_data := (others => '0');
elsif tlb_load = '1' then
addr := r.addr(63 downto 12) & x"000";
tlb_data := pte;
elsif ptbl_rd = '1' then
addr := x"00" & r.ptcr(55 downto 12) & x"008";
tlb_data := (others => '0');
elsif prtbl_rd = '1' then
addr := prtable_addr;
tlb_data := (others => '0');
else
addr := pgtable_addr;
tlb_data := (others => '0');
end if;
l_out.done <= r.done;
l_out.err <= r.err;
l_out.invalid <= r.invalid;
l_out.badtree <= r.badtree;
l_out.segerr <= r.segerror;
l_out.perm_error <= r.perm_err;
l_out.rc_error <= r.rc_error;
d_out.valid <= dcreq;
d_out.tlbie <= tlbie_req;
d_out.doall <= r.inval_all;
d_out.tlbld <= tlb_load;
d_out.addr <= addr;
d_out.pte <= tlb_data;
i_out.tlbld <= itlb_load;
i_out.tlbie <= tlbie_req;
i_out.doall <= r.inval_all;
i_out.addr <= addr;
i_out.pte <= tlb_data;
end process;
end;
View Git Blame Copy Permalink