You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
microwatt/multiply.vhdl

127 lines
4.0 KiB
VHDL

library ieee;
use ieee.std_logic_1164.all;
use ieee.numeric_std.all;
library work;
use work.common.all;
use work.decode_types.all;
entity multiply is
generic (
PIPELINE_DEPTH : natural := 16
);
port (
clk : in std_logic;
m_in : in Execute1ToMultiplyType;
m_out : out MultiplyToExecute1Type
);
end entity multiply;
architecture behaviour of multiply is
signal m: Execute1ToMultiplyType;
type multiply_pipeline_stage is record
valid : std_ulogic;
insn_type : insn_type_t;
data : signed(129 downto 0);
write_reg : std_ulogic_vector(4 downto 0);
rc : std_ulogic;
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
oe : std_ulogic;
is_32bit : std_ulogic;
xerc : xer_common_t;
end record;
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
constant MultiplyPipelineStageInit : multiply_pipeline_stage := (valid => '0',
insn_type => OP_ILLEGAL,
rc => '0', oe => '0',
is_32bit => '0',
xerc => xerc_init,
data => (others => '0'),
others => (others => '0'));
type multiply_pipeline_type is array(0 to PIPELINE_DEPTH-1) of multiply_pipeline_stage;
constant MultiplyPipelineInit : multiply_pipeline_type := (others => MultiplyPipelineStageInit);
type reg_type is record
multiply_pipeline : multiply_pipeline_type;
end record;
signal r, rin : reg_type := (multiply_pipeline => MultiplyPipelineInit);
begin
multiply_0: process(clk)
begin
if rising_edge(clk) then
m <= m_in;
r <= rin;
end if;
end process;
multiply_1: process(all)
variable v : reg_type;
variable d : std_ulogic_vector(129 downto 0);
variable d2 : std_ulogic_vector(63 downto 0);
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
variable ov : std_ulogic;
begin
v := r;
m_out <= MultiplyToExecute1Init;
v.multiply_pipeline(0).valid := m.valid;
v.multiply_pipeline(0).insn_type := m.insn_type;
v.multiply_pipeline(0).data := signed(m.data1) * signed(m.data2);
v.multiply_pipeline(0).write_reg := m.write_reg;
v.multiply_pipeline(0).rc := m.rc;
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
v.multiply_pipeline(0).oe := m.oe;
v.multiply_pipeline(0).is_32bit := m.is_32bit;
v.multiply_pipeline(0).xerc := m.xerc;
loop_0: for i in 1 to PIPELINE_DEPTH-1 loop
v.multiply_pipeline(i) := r.multiply_pipeline(i-1);
end loop;
d := std_ulogic_vector(v.multiply_pipeline(PIPELINE_DEPTH-1).data);
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
ov := '0';
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
-- TODO: Handle overflows
case_0: case v.multiply_pipeline(PIPELINE_DEPTH-1).insn_type is
when OP_MUL_L64 =>
d2 := d(63 downto 0);
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
if v.multiply_pipeline(PIPELINE_DEPTH-1).is_32bit = '1' then
ov := (or d(63 downto 31)) and not (and d(63 downto 31));
else
ov := (or d(127 downto 63)) and not (and d(127 downto 63));
end if;
when OP_MUL_H32 =>
d2 := d(63 downto 32) & d(63 downto 32);
when OP_MUL_H64 =>
d2 := d(127 downto 64);
when others =>
--report "Illegal insn type in multiplier";
d2 := (others => '0');
end case;
m_out.write_reg_data <= d2;
m_out.write_reg_nr <= v.multiply_pipeline(PIPELINE_DEPTH-1).write_reg;
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
m_out.xerc <= v.multiply_pipeline(PIPELINE_DEPTH-1).xerc;
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
-- Generate OV/OV32/SO when OE=1
if v.multiply_pipeline(PIPELINE_DEPTH-1).valid = '1' then
m_out.valid <= '1';
m_out.rc <= v.multiply_pipeline(PIPELINE_DEPTH-1).rc;
Add basic XER support The carry is currently internal to execute1. We don't handle any of the other XER fields. This creates type called "xer_common_t" that contains the commonly used XER bits (CA, CA32, SO, OV, OV32). The value is stored in the CR file (though it could be a separate module). The rest of the bits will be implemented as a separate SPR and the two parts reconciled in mfspr/mtspr in latter commits. We always read XER in decode2 (there is little point not to) and send it down all pipeline branches as it will be needed in writeback for all type of instructions when CR0:SO needs to be updated (such forms exist for all pipeline branches even if we don't yet implement them). To avoid having to track XER hazards, we forward it back in EX1. This assumes that other pipeline branches that can modify it (mult and div) are running single issue for now. One additional hazard to beware of is an XER:SO modifying instruction in EX1 followed immediately by a store conditional. Due to our writeback latency, the store will go down the LSU with the previous XER value, thus the stcx. will set CR0:SO using an obsolete SO value. I doubt there exist any code relying on this behaviour being correct but we should account for it regardless, possibly by ensuring that stcx. remain single issue initially, or later by adding some minimal tracking or moving the LSU into the same pipeline as execute. Missing some obscure XER affecting instructions like addex or mcrxrx. [paulus@ozlabs.org - fix CA32 and OV32 for OP_ADD, fix order of arguments to set_ov] Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
5 years ago
m_out.write_xerc_enable <= v.multiply_pipeline(PIPELINE_DEPTH-1).oe;
-- We must test oe because the RC update code in writeback
-- will use the xerc value to set CR0:SO so we must not clobber
-- xerc if OE wasn't set.
--
if v.multiply_pipeline(PIPELINE_DEPTH-1).oe = '1' then
m_out.xerc.ov <= ov;
m_out.xerc.ov32 <= ov;
m_out.xerc.so <= v.multiply_pipeline(PIPELINE_DEPTH-1).xerc.so or ov;
end if;
end if;
rin <= v;
end process;
end architecture behaviour;