head 1.10; access; symbols perseant-exfatfs-base-20250801:1.10 perseant-exfatfs-base-20240630:1.10 perseant-exfatfs:1.10.0.6 perseant-exfatfs-base:1.10 netbsd-8-3-RELEASE:1.9 cjep_sun2x:1.10.0.4 cjep_sun2x-base:1.10 cjep_staticlib_x-base1:1.10 cjep_staticlib_x:1.10.0.2 cjep_staticlib_x-base:1.10 phil-wifi-20200421:1.10 phil-wifi-20200411:1.10 phil-wifi-20200406:1.10 netbsd-8-2-RELEASE:1.9 netbsd-8-1-RELEASE:1.9 netbsd-8-1-RC1:1.9 pgoyette-compat-merge-20190127:1.9.66.1 pgoyette-compat-20190127:1.10 pgoyette-compat-20190118:1.10 pgoyette-compat-1226:1.10 pgoyette-compat-1126:1.10 pgoyette-compat-1020:1.10 pgoyette-compat-0930:1.10 pgoyette-compat-0906:1.10 netbsd-7-2-RELEASE:1.9 pgoyette-compat-0728:1.10 netbsd-8-0-RELEASE:1.9 phil-wifi:1.9.0.68 phil-wifi-base:1.9 pgoyette-compat-0625:1.9 netbsd-8-0-RC2:1.9 pgoyette-compat-0521:1.9 pgoyette-compat-0502:1.9 pgoyette-compat-0422:1.9 netbsd-8-0-RC1:1.9 pgoyette-compat-0415:1.9 pgoyette-compat-0407:1.9 pgoyette-compat-0330:1.9 pgoyette-compat-0322:1.9 pgoyette-compat-0315:1.9 netbsd-7-1-2-RELEASE:1.9 pgoyette-compat:1.9.0.66 pgoyette-compat-base:1.9 netbsd-7-1-1-RELEASE:1.9 matt-nb8-mediatek:1.9.0.64 matt-nb8-mediatek-base:1.9 perseant-stdc-iso10646:1.9.0.62 perseant-stdc-iso10646-base:1.9 netbsd-8:1.9.0.60 netbsd-8-base:1.9 prg-localcount2-base3:1.9 prg-localcount2-base2:1.9 prg-localcount2-base1:1.9 prg-localcount2:1.9.0.58 prg-localcount2-base:1.9 pgoyette-localcount-20170426:1.9 bouyer-socketcan-base1:1.9 pgoyette-localcount-20170320:1.9 netbsd-7-1:1.9.0.56 netbsd-7-1-RELEASE:1.9 netbsd-7-1-RC2:1.9 netbsd-7-nhusb-base-20170116:1.9 bouyer-socketcan:1.9.0.54 bouyer-socketcan-base:1.9 pgoyette-localcount-20170107:1.9 netbsd-7-1-RC1:1.9 pgoyette-localcount-20161104:1.9 netbsd-7-0-2-RELEASE:1.9 localcount-20160914:1.9 netbsd-7-nhusb:1.9.0.52 netbsd-7-nhusb-base:1.9 pgoyette-localcount-20160806:1.9 pgoyette-localcount-20160726:1.9 pgoyette-localcount:1.9.0.50 pgoyette-localcount-base:1.9 netbsd-7-0-1-RELEASE:1.9 netbsd-7-0:1.9.0.48 netbsd-7-0-RELEASE:1.9 netbsd-7-0-RC3:1.9 netbsd-7-0-RC2:1.9 netbsd-7-0-RC1:1.9 netbsd-5-2-3-RELEASE:1.9 netbsd-5-1-5-RELEASE:1.9 netbsd-6-0-6-RELEASE:1.9 netbsd-6-1-5-RELEASE:1.9 netbsd-7:1.9.0.46 netbsd-7-base:1.9 yamt-pagecache-base9:1.9 yamt-pagecache-tag8:1.9 netbsd-6-1-4-RELEASE:1.9 netbsd-6-0-5-RELEASE:1.9 tls-earlyentropy:1.9.0.44 tls-earlyentropy-base:1.9 riastradh-xf86-video-intel-2-7-1-pre-2-21-15:1.9 riastradh-drm2-base3:1.9 netbsd-6-1-3-RELEASE:1.9 netbsd-6-0-4-RELEASE:1.9 netbsd-5-2-2-RELEASE:1.9 netbsd-5-1-4-RELEASE:1.9 netbsd-6-1-2-RELEASE:1.9 netbsd-6-0-3-RELEASE:1.9 netbsd-5-2-1-RELEASE:1.9 netbsd-5-1-3-RELEASE:1.9 netbsd-6-1-1-RELEASE:1.9 riastradh-drm2-base2:1.9 riastradh-drm2-base1:1.9 riastradh-drm2:1.9.0.38 riastradh-drm2-base:1.9 netbsd-6-1:1.9.0.42 netbsd-6-0-2-RELEASE:1.9 netbsd-6-1-RELEASE:1.9 netbsd-6-1-RC4:1.9 netbsd-6-1-RC3:1.9 agc-symver:1.9.0.40 agc-symver-base:1.9 netbsd-6-1-RC2:1.9 netbsd-6-1-RC1:1.9 yamt-pagecache-base8:1.9 netbsd-5-2:1.9.0.36 netbsd-6-0-1-RELEASE:1.9 yamt-pagecache-base7:1.9 netbsd-5-2-RELEASE:1.9 netbsd-5-2-RC1:1.9 matt-nb6-plus-nbase:1.9 yamt-pagecache-base6:1.9 netbsd-6-0:1.9.0.34 netbsd-6-0-RELEASE:1.9 netbsd-6-0-RC2:1.9 tls-maxphys:1.9.0.32 tls-maxphys-base:1.9 matt-nb6-plus:1.9.0.30 matt-nb6-plus-base:1.9 netbsd-6-0-RC1:1.9 yamt-pagecache-base5:1.9 yamt-pagecache-base4:1.9 netbsd-6:1.9.0.28 netbsd-6-base:1.9 netbsd-5-1-2-RELEASE:1.9 netbsd-5-1-1-RELEASE:1.9 yamt-pagecache-base3:1.9 yamt-pagecache-base2:1.9 yamt-pagecache:1.9.0.26 yamt-pagecache-base:1.9 cherry-xenmp:1.9.0.24 cherry-xenmp-base:1.9 bouyer-quota2-nbase:1.9 bouyer-quota2:1.9.0.22 bouyer-quota2-base:1.9 matt-mips64-premerge-20101231:1.9 matt-nb5-mips64-premerge-20101231:1.9 matt-nb5-pq3:1.9.0.20 matt-nb5-pq3-base:1.9 netbsd-5-1:1.9.0.18 netbsd-5-1-RELEASE:1.9 netbsd-5-1-RC4:1.9 matt-nb5-mips64-k15:1.9 netbsd-5-1-RC3:1.9 netbsd-5-1-RC2:1.9 netbsd-5-1-RC1:1.9 netbsd-5-0-2-RELEASE:1.9 matt-nb5-mips64-premerge-20091211:1.9 matt-premerge-20091211:1.9 matt-nb5-mips64-u2-k2-k4-k7-k8-k9:1.9 matt-nb4-mips64-k7-u2a-k9b:1.9 matt-nb5-mips64-u1-k1-k5:1.9 matt-nb5-mips64:1.9.0.16 netbsd-5-0-1-RELEASE:1.9 jym-xensuspend-nbase:1.9 netbsd-5-0:1.9.0.14 netbsd-5-0-RELEASE:1.9 netbsd-5-0-RC4:1.9 netbsd-5-0-RC3:1.9 netbsd-5-0-RC2:1.9 jym-xensuspend:1.9.0.12 jym-xensuspend-base:1.9 netbsd-5-0-RC1:1.9 netbsd-5:1.9.0.10 netbsd-5-base:1.9 matt-mips64-base2:1.9 mjf-devfs2:1.9.0.8 mjf-devfs2-base:1.9 wrstuden-revivesa-base-3:1.9 wrstuden-revivesa-base-2:1.9 wrstuden-revivesa-base-1:1.9 yamt-pf42-base4:1.9 yamt-pf42-base3:1.9 hpcarm-cleanup-nbase:1.9 hpcarm-cleanup-base:1.9 yamt-pf42-baseX:1.9 yamt-pf42-base2:1.9 wrstuden-revivesa:1.9.0.6 wrstuden-revivesa-base:1.9 yamt-pf42:1.9.0.4 yamt-pf42-base:1.9 keiichi-mipv6-nbase:1.9 keiichi-mipv6:1.9.0.2 keiichi-mipv6-base:1.9 matt-armv6-base:1.5 matt-armv6:1.5.0.2 matt-armv6-nbase:1.9; locks; strict; comment @# @; 1.10 date 2018.07.13.07.56.29; author maxv; state dead; branches; next 1.9; commitid gYDOFMQmW7CLSWJA; 1.9 date 2008.01.30.14.16.42; author ad; state Exp; branches 1.9.66.1 1.9.68.1; next 1.8; 1.8 date 2008.01.14.12.48.52; author yamt; state Exp; branches; next 1.7; 1.7 date 2008.01.14.12.47.59; author yamt; state Exp; branches; next 1.6; 1.6 date 2008.01.14.12.42.02; author yamt; state Exp; branches; next 1.5; 1.5 date 2008.01.05.22.54.49; author yamt; state Exp; branches 1.5.2.1; next 1.4; 1.4 date 2008.01.05.22.49.59; author yamt; state Exp; branches; next 1.3; 1.3 date 2008.01.02.01.17.41; author yamt; state Exp; branches; next 1.2; 1.2 date 2008.01.02.01.06.31; author yamt; state Exp; branches; next 1.1; 1.1 date 2008.01.01.23.42.56; author yamt; state Exp; branches; next ; 1.9.66.1 date 2018.07.28.04.38.15; author pgoyette; state dead; branches; next ; commitid 1UP1xAIUxv1ZgRLA; 1.9.68.1 date 2019.06.10.22.10.43; author christos; state dead; branches; next ; commitid jtc8rnCzWiEEHGqB; 1.5.2.1 date 2008.01.05.22.54.49; author matt; state dead; branches; next 1.5.2.2; 1.5.2.2 date 2008.01.09.02.02.34; author matt; state Exp; branches; next 1.5.2.3; 1.5.2.3 date 2008.03.23.00.50.08; author matt; state Exp; branches; next ; desc @@ 1.10 log @Revamp tprof. Rewrite the Intel backend to use the generic PMC interface, which is available on all Intel CPUs. Synchronize the AMD backend with the new interface. The kernel identifies the PMC interface, and gives its id to userland. Userland then queries the events itself (via cpuid etc). These events depend on the PMC interface. The tprof utility is rewritten to allow the user to choose which event to count (which was not possible until now, the event was hardcoded in the backend). The command line format is based on usr.bin/pmc, eg: tprof -e llc-misses:k -o output sleep 20 The man page is updated too, but the arguments will likely change soon anyway so it doesn't matter a lot. The tprof utility has three tables: Intel Architectural Version 1 Intel Skylake/Kabylake AMD Family 10h A CPU can support a combination of tables. For example Kabylake has Intel-Architectural-Version-1 and its own Intel-Kabylake table. For now the Intel Skylake/Kabylake table contains only one event, just to demonstrate that the combination of tables works. Tested on an Intel Core i5 Kabylake. The code for AMD Family 10h is taken from the code I had written for usr.bin/pmc. I haven't tested it yet, but it's the same as pmc(1), so I guess it works as-is. The whole thing is written in such a way that (I think) it is not complicated to add more CPU models, and more architectures (other than x86). @ text @$NetBSD: README,v 1.9 2008/01/30 14:16:42 ad Exp $ NOTE: - tprof driver currently only supports pentium4 (netburst) processors. - it samples program counters on every PMIs. - it's currently hardcoded to use global_power_events events. for details, see x86/x86/tprof_pmi.c and intel's processor manuals. usage: 0. set SIZEOF_PTR environment variable, which is used by tpfmt.sh and tpann.sh. if not set, SIZEOF_PTR=4 is assumed. 1. add a line to your kernel config. pseudo-device tprof 2. create a device special file. # mknod /dev/tprof c 191 0 3. run the tprof command. # tprof -o /tmp/foo sleep 1 tprof statistics: sample 57 overflow 0 buf 3 emptybuf 3 dropbuf 0 dropbuf_sample 0 4. format the result. the first line in the following example means that 11 samples have been taken at 0xc0396c36, whose symbolic name is lapic_gettick+0x6. # sh ./tpfmt.sh < /tmp/foo 11 c0396c36 lapic_gettick+0x6 5 c039b98a x86_pause+0x2 4 c010cf9d __cpu_simple_lock+0xd 2 c010cfcd __cpu_simple_lock_try+0xd 2 c039b571 bus_space_read_4+0x11 1 c01005c8 sse2_zero_page+0x18 1 c0100624 sse2_copy_page+0x34 1 c010ceeb mutex_spin_enter+0x2b 1 c010cef5 mutex_spin_enter+0x35 1 c010cf32 mutex_spin_exit+0x32 1 c0119ed0 in_localaddr+0x30 1 c012d0fd tcp_output+0x1fbd 1 c02980c2 amap_copy+0x42 1 c02a0100 uvm_map_lookup_entry_bytree+0x20 1 c02a27fe uvm_tree_RB_REMOVE+0xee 1 c02a8914 uvm_pagelookup+0x4 1 c02a9d5c uvm_pagefree+0xfc 1 c02a9e36 uvm_pagefree+0x1d6 1 c02dd9d1 _kernel_unlock+0xa1 1 c02e0285 mutex_vector_enter+0x15 1 c02eb83a sleepq_wake+0x5a 1 c0303467 pool_cache_get_paddr+0x97 1 c030368b pool_cache_put_slow+0x6b 1 c0321ed3 pffasttimo+0x33 1 c034547a VOP_LOCK+0xa 1 c0346235 VOP_ACCESS+0x45 1 c034a749 genfs_unlock+0x29 1 c038f251 cpu_idle+0x31 1 c03938da pmap_write_protect+0xaa 1 c0394305 pmap_do_remove+0x2e5 1 c03944b3 pmap_do_remove+0x493 1 c0396cdf lapic_delay+0x5f 1 c0396d19 lapic_delay+0x99 1 c0396d1d lapic_delay+0x9d 1 c0397429 lapic_clockintr+0x19 1 c039b984 x86_mwait+0xc 1 c042f66a _atomic_swap_32+0xa 5. tpann.sh is another formatter. it outputs "objdump -d" with numbers of samples for each addresses. # tprof -o /tmp/bar sleep 100 # sh ./tpann.sh < /tmp/bar : snip : c01005e0 : 4 c01005e0: 55 push %ebp 11 c01005e1: 89 e5 mov %esp,%ebp 1 c01005e3: 8b 54 24 08 mov 0x8(%esp),%edx 3 c01005e7: b9 00 10 00 00 mov $0x1000,%ecx 1 c01005ec: 31 c0 xor %eax,%eax 1 c01005ee: 89 f6 mov %esi,%esi 7936 c01005f0: 0f c3 42 00 movnti %eax,0x0(%edx) 6371 c01005f4: 0f c3 42 04 movnti %eax,0x4(%edx) 1220 c01005f8: 0f c3 42 08 movnti %eax,0x8(%edx) 741 c01005fc: 0f c3 42 0c movnti %eax,0xc(%edx) 1178 c0100600: 0f c3 42 10 movnti %eax,0x10(%edx) 1334 c0100604: 0f c3 42 14 movnti %eax,0x14(%edx) 976 c0100608: 0f c3 42 18 movnti %eax,0x18(%edx) 1299 c010060c: 0f c3 42 1c movnti %eax,0x1c(%edx) 954 c0100610: 83 e9 20 sub $0x20,%ecx 45 c0100613: 8d 52 20 lea 0x20(%edx),%edx 238 c0100616: 75 d8 jne c01005f0 71 c0100618: 0f ae f8 sfence 297 c010061b: 5d pop %ebp 19 c010061c: c3 ret 0 c010061d: 8d 76 00 lea 0x0(%esi),%esi : snip : @ 1.9 log @Expunge references to lockmgr. @ text @d1 1 a1 1 $NetBSD: README,v 1.8 2008/01/14 12:48:52 yamt Exp $ @ 1.9.68.1 log @Sync with HEAD @ text @d1 1 a1 1 $NetBSD: README,v 1.9 2008/01/30 14:16:42 ad Exp $ @ 1.9.66.1 log @Sync with HEAD @ text @d1 1 a1 1 $NetBSD: README,v 1.9 2008/01/30 14:16:42 ad Exp $ @ 1.8 log @whitespace @ text @d1 1 a1 1 $NetBSD: README,v 1.7 2008/01/14 12:47:59 yamt Exp $ a56 1 1 c02dd4f6 lockmgr+0x3d6 @ 1.7 log @add a script to produce something like opannotate. @ text @d1 1 a1 1 $NetBSD: README,v 1.6 2008/01/14 12:42:02 yamt Exp $ d15 1 d19 1 d23 1 d37 1 d80 1 @ 1.6 log @rename fmt.sh to tpfmt.sh @ text @d1 1 a1 1 $NetBSD: README,v 1.5 2008/01/05 22:54:49 yamt Exp $ d11 3 a33 2 (tpfmt.sh uses the SIZEOF_PTR environment variable. if not set, SIZEOF_PTR=4 is assumed.) d73 36 @ 1.5 log @mention SIZEOF_PTR @ text @d1 1 a1 1 $NetBSD: README,v 1.4 2008/01/05 22:49:59 yamt Exp $ d31 1 a31 1 (fmt.sh uses the SIZEOF_PTR environment variable. d33 1 a33 1 # sh ./fmt.sh < /tmp/foo @ 1.5.2.1 log @file README was added on branch matt-armv6 on 2008-01-09 02:02:34 +0000 @ text @d1 71 @ 1.5.2.2 log @sync with HEAD @ text @a0 71 $NetBSD$ NOTE: - tprof driver currently only supports pentium4 (netburst) processors. - it samples program counters on every PMIs. - it's currently hardcoded to use global_power_events events. for details, see x86/x86/tprof_pmi.c and intel's processor manuals. usage: 1. add a line to your kernel config. pseudo-device tprof 2. create a device special file. # mknod /dev/tprof c 191 0 3. run the tprof command. # tprof -o /tmp/foo sleep 1 tprof statistics: sample 57 overflow 0 buf 3 emptybuf 3 dropbuf 0 dropbuf_sample 0 4. format the result. the first line in the following example means that 11 samples have been taken at 0xc0396c36, whose symbolic name is lapic_gettick+0x6. (fmt.sh uses the SIZEOF_PTR environment variable. if not set, SIZEOF_PTR=4 is assumed.) # sh ./fmt.sh < /tmp/foo 11 c0396c36 lapic_gettick+0x6 5 c039b98a x86_pause+0x2 4 c010cf9d __cpu_simple_lock+0xd 2 c010cfcd __cpu_simple_lock_try+0xd 2 c039b571 bus_space_read_4+0x11 1 c01005c8 sse2_zero_page+0x18 1 c0100624 sse2_copy_page+0x34 1 c010ceeb mutex_spin_enter+0x2b 1 c010cef5 mutex_spin_enter+0x35 1 c010cf32 mutex_spin_exit+0x32 1 c0119ed0 in_localaddr+0x30 1 c012d0fd tcp_output+0x1fbd 1 c02980c2 amap_copy+0x42 1 c02a0100 uvm_map_lookup_entry_bytree+0x20 1 c02a27fe uvm_tree_RB_REMOVE+0xee 1 c02a8914 uvm_pagelookup+0x4 1 c02a9d5c uvm_pagefree+0xfc 1 c02a9e36 uvm_pagefree+0x1d6 1 c02dd4f6 lockmgr+0x3d6 1 c02dd9d1 _kernel_unlock+0xa1 1 c02e0285 mutex_vector_enter+0x15 1 c02eb83a sleepq_wake+0x5a 1 c0303467 pool_cache_get_paddr+0x97 1 c030368b pool_cache_put_slow+0x6b 1 c0321ed3 pffasttimo+0x33 1 c034547a VOP_LOCK+0xa 1 c0346235 VOP_ACCESS+0x45 1 c034a749 genfs_unlock+0x29 1 c038f251 cpu_idle+0x31 1 c03938da pmap_write_protect+0xaa 1 c0394305 pmap_do_remove+0x2e5 1 c03944b3 pmap_do_remove+0x493 1 c0396cdf lapic_delay+0x5f 1 c0396d19 lapic_delay+0x99 1 c0396d1d lapic_delay+0x9d 1 c0397429 lapic_clockintr+0x19 1 c039b984 x86_mwait+0xc 1 c042f66a _atomic_swap_32+0xa @ 1.5.2.3 log @sync with HEAD @ text @d1 1 a1 1 README,v 1.5.2.2 2008/01/09 02:02:34 matt Exp a10 3 0. set SIZEOF_PTR environment variable, which is used by tpfmt.sh and tpann.sh. if not set, SIZEOF_PTR=4 is assumed. a11 1 a14 1 a17 1 d31 3 a33 2 # sh ./tpfmt.sh < /tmp/foo d52 1 a71 37 5. tpann.sh is another formatter. it outputs "objdump -d" with numbers of samples for each addresses. # tprof -o /tmp/bar sleep 100 # sh ./tpann.sh < /tmp/bar : snip : c01005e0 : 4 c01005e0: 55 push %ebp 11 c01005e1: 89 e5 mov %esp,%ebp 1 c01005e3: 8b 54 24 08 mov 0x8(%esp),%edx 3 c01005e7: b9 00 10 00 00 mov $0x1000,%ecx 1 c01005ec: 31 c0 xor %eax,%eax 1 c01005ee: 89 f6 mov %esi,%esi 7936 c01005f0: 0f c3 42 00 movnti %eax,0x0(%edx) 6371 c01005f4: 0f c3 42 04 movnti %eax,0x4(%edx) 1220 c01005f8: 0f c3 42 08 movnti %eax,0x8(%edx) 741 c01005fc: 0f c3 42 0c movnti %eax,0xc(%edx) 1178 c0100600: 0f c3 42 10 movnti %eax,0x10(%edx) 1334 c0100604: 0f c3 42 14 movnti %eax,0x14(%edx) 976 c0100608: 0f c3 42 18 movnti %eax,0x18(%edx) 1299 c010060c: 0f c3 42 1c movnti %eax,0x1c(%edx) 954 c0100610: 83 e9 20 sub $0x20,%ecx 45 c0100613: 8d 52 20 lea 0x20(%edx),%edx 238 c0100616: 75 d8 jne c01005f0 71 c0100618: 0f ae f8 sfence 297 c010061b: 5d pop %ebp 19 c010061c: c3 ret 0 c010061d: 8d 76 00 lea 0x0(%esi),%esi : snip : @ 1.4 log @PR/37662 was fixed. @ text @d1 1 a1 1 $NetBSD: README,v 1.3 2008/01/02 01:17:41 yamt Exp $ d31 2 @ 1.3 log @note what's sampled. @ text @d1 1 a1 1 $NetBSD: README,v 1.2 2008/01/02 01:06:31 yamt Exp $ a7 1 - it doesn't work for amd64 due to a bug. see PR/37662. @ 1.2 log @explain the output of fmt.sh @ text @d1 1 a1 1 $NetBSD: README,v 1.1 2008/01/01 23:42:56 yamt Exp $ d5 3 @ 1.1 log @minimum documentation. @ text @d1 1 a1 1 $NetBSD$ d27 2 @