Apple M1 Microarchitecture Research by Dougall Johnson

Firestorm: Overview | Base Instructions | SIMD and FP Instructions
Icestorm:  Overview | Base Instructions | SIMD and FP Instructions

SQSHRUN (2D)

Test 1: uops

Code:

  sqshrun v0.2s, v0.2d, #3
  movi v0.16b, 1
  movi v1.16b, 2

(no loop instructions)

1000 unrolls and 1 iteration

Retires: 1.000

Issues: 1.000

Integer unit issues: 0.000

Load/store unit issues: 0.000

SIMD/FP unit issues: 1.000

retire (01)cycle (02)0318191e3f4e51inst issue (52)~issue fp/simd (54)~dispatch fp/simd (57)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op fp/simd (7e)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst neon or fp (9a)a8cfd5d6ddinst fetch restart (de)e0? fp/simd (ee)f5f6f7f8fd
1004303723000612547251000100010003981600301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
1004303723000612547251000100010003981600301830373037241432895100010001000303730371110011000173116112629100030383038303830383038
1004303723000612547251000100010003981600301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
1004303723000612547251000100010003981600301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
1004303723000822547251000100010003981600301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
10043037221012612547251000100010003981600301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
1004303722000612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
1004303723000612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
1004303722000612547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038
10043037230001452547251000100010003981601301830373037241432895100010001000303730371110011000073116112629100030383038303830383038

Test 2: Latency 1->2

Code:

  sqshrun v0.2s, v0.2d, #3
  movi v0.16b, 1
  movi v1.16b, 2

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)03080b18191e1f3a3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)inst barrier (9c)9fa1a6a7a8a9acc2cdcfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
102043003722500000006129547251010010010000100100005004277160030018300373003728264252874510100200100002001000030037300371110201100991001001000001000000000071021622296330100001003003830038300383003830038
10204300372250000000612954725101001001000010010000500427716003001830037300372826432874510100200100002001000030037300371110201100991001001000001000000000071021622296330100001003003830038300383003830038
102043003722400000007482954725101001001000010010000500427716003001830037300372826432874510100200100002001000030037300371110201100991001001000001000000000071021622296330100001003003830038300383003830038
10204300372250000000612954725101221001000010010150500427716003001830037300372826430288531104322011159222111533037130323811020110099100100100000100000121943320868372542986831100001003037130419303703037030362
10204303232270088936616045752949315910183139100161391105069642866240302703037330367282902628838111912251116422411154304183037481102011009910010010000010020000000710216222963330100001003003830227301813027230230
1020430179227116455844004653295021011017415210080132106006694289328030018302783013128264328745105842081000020610000301323022711102011009910010010000110022200296300071021622296330100001003018130181301803017930182
10204301312341121088113932945643101001001000010010000649427716013001830037300372826932874510100200100002001000030037300851110201100991001001000001000000000074124822296660100001003003830038300383003830038
10204300372250000000612954725101001001000010010000500427716013001830037300372826432874510100200100002001000030037300371110201100991001001000001000000000071021622296330100001003003830038300383003830038
10204300372250000000612954725101001001000010010000500427716003001830037300372826432874510100200100002001000030037300371110201100991001001000001000000000071021622296330100001003003830038300383003830038
102043003722500000002512954725101001001000010010000500427716013001830037300372826432874510100200100002001000030037300371110201100991001001000001000000000071021622296330100001003003830038300383003830038

1000 unrolls and 10 iterations

Result (median cycles for code): 3.0037

retire (01)cycle (02)030b18191e3f4e51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa8cfd5d6ddinst fetch restart (de)e0? int output thing (e9)? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
10024300372250000726295472510010101000010100005042771600300183003730037282863287671001020100002010000300373003711100211091010100001006408163429629010000103003830038300383003830038
10024300372240000726295472510010101000010100005042771600300183003730037282863287671001020100002010000300373003711100211091010100001006402164429629010000103003830038300383003830038
1002430037225000061295472510010101000010100005042771600300183003730037282863287671001020100002010000300373003711100211091010100001006404163329629010000103003830038300383003830038
1002430037225000061295472510010101000010100005042771600300183003730037282863287671001020100002010000300373003711100211091010100001006404164429629010000103003830038300383003830038
1002430037224000061295472510010101000010100005042771600300183003730037282863287671001020100002010000300373003711100211091010100001006404163429629010000103003830038300383003830038
1002430037225000061295472510010101000010100005042771600300183003730037282863287671001020100002010000300373003711100211091010100001006404414429629010000103003830038300383003830038
100243003722400005683295472510010101000010100005042771600300183003730037282863287671001020100002010000300373003711100211091010100001006403164329629010000103003830038300383003830038
1002430037225000061295472510010101000010100005042771601300183003730037282863287671001020100002010000300373003711100211091010100001006404164429629010000103003830038300383003830038
1002430037225000061295472510010101000010100005042771600300183003730037282863287671001020100002010000300373003711100211091010100001006404163329629010000103003830038300383003830038
10024300372250001261295472510010101000010100005042771600300183003730037282863287671001020100002010000300373003711100211091010100001006403163429629010000103003830038300383003830038

Test 3: throughput

Count: 8

Code:

  sqshrun v0.2s, v8.2d, #3
  sqshrun v1.2s, v8.2d, #3
  sqshrun v2.2s, v8.2d, #3
  sqshrun v3.2s, v8.2d, #3
  sqshrun v4.2s, v8.2d, #3
  sqshrun v5.2s, v8.2d, #3
  sqshrun v6.2s, v8.2d, #3
  sqshrun v7.2s, v8.2d, #3
  movi v8.16b, 9

(fused SUBS/B.cc loop)

100 unrolls and 100 iterations

Result (median cycles for code divided by count): 0.2505

retire (01)cycle (02)031e3f51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)60696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa8acc5branch mispredict (cb)cdcfd5d6inst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8020420039186030258010810080008100800205006401320200202003920039997769990801202008003220080032200392003911802011009910010080000100001115118216020036800001002004020040200402004020040
8020420039186930258010810080008100800205006401321200202003920039997769990801202008003220080032200392003911802011009910010080000100101115118016020036800001002004020040200402004020040
802042003918603025801081008000810080020500640132020020200392003999776999080120200800322008003220039200391180201100991001008000010001381115118016020036800001002004020040200402004020040
8020420039185030258010810080008100800205006401320200202003920039997769990801202008003220080032200392003911802011009910010080000100001115154016020036800001002004020040200402004020040
8020420039173030258010810080008100800205006401320200202003920039997769990801202008003220080032200392003911802011009910010080000100001115118016020036800001002004020040200402004020040
8020420039173030258010810080008100800205006401320200202003920039997769990801202008003220080032200392003911802011009910010080000100001115118016020036800001002004020040200402004020040
8020420039173030258010810080008100800205006401320200202003920039997769990801202008003220080032200392003911802011009910010080000100001115118016020036800001002004020040200402004020040
8020420039174030258010810080008100800205006401321200202003920039997769990801202008003220080032200392003911802011009910010080000100001115118016020036800001002004020040200402004020040
8020420039161030258010810080008100800205006401320200202003920039997769990801202008003220080032200392003911802011009910010080000100001115118025020036800001002004020040200402004020040
8020420039161030258010810080008100800205006401320200202003920039997769990801202008003220080032200392003911802011009910010080000100001115118016020036800001002004020040200402004020040

1000 unrolls and 10 iterations

Result (median cycles for code divided by count): 0.2505

retire (01)cycle (02)030b1e1f3f51inst issue (52)~issue int (53)~issue fp/simd (54)~dispatch int (56)~dispatch fp/simd (57)huge thing int (59)huge thing fp/simd (5b)696d6edispatch stall (70)scheduler rewind (75)scheduler stall (76)~dispatch op (78)~map op int (7c)~map op fp/simd (7e)~map lookup int (7f)~map lookup fp/simd (81)8283pipeline redirect (84)85inst all (8c)inst branch (8d)inst branch taken (90)inst b.cc (94)inst integer (97)inst neon or fp (9a)9fa1a8accfd5d6ddinst fetch restart (de)e0? fp/simd (ee)gpr retires (ef)f5f6f7f8fd
8002420050151000230258001010800001080000506400002002020039200399996310019800102080000208000020039200391180021109101080000100005020416342003680000102004020040200402004020040
800242003915000040258001010800001080000506400002002020039200399996310019800102080000208000020039200391180021109101080000100005020716762003680000102004020040200402004020040
800242003915000040258001010800001080000506400002002020039200399996310019800102080000208000020039200391180021109101080000100005020616462003680000102004020040200402004020040
80024200391500004025800101080000108000050640000200202011520195999631001980010208000020800002003920039118002110910108000010001445020716672003680000102004020040200402004020040
800242003915000040258001010800001080000506400002002020039200399996310019800102080000208000020039200391180021109101080000100005020616642003680000102004020040200402004020040
800242003915000040258001010800001080000506400002002020039200399996310019800102080000208000020039200391180021109101080000100005020616342003680000102004020040200402004020040
800242003915000040258001010800001080000506400002002020039200399996310019800102080000208000020039200391180021109101080000100005020616762003680000102004020040200402004020040
800242003915000040258001010800001080000506400002002020039200399996310019800102080000208000020039200391180021109101080000100095020416452003680000102004020040200402004020040
800242003915000040258001010800001080000506400002002020039200399996310019800102080000208000020039200391180021109101080000100005020416482003680000102004020040200402004020040
800242003915000040258001010800001080000506400002002020039200399996310019800102080000208000020039200391180021109101080000100005020716662003680000102004020040200402004020040