Thread (5 messages) 5 messages, 2 authors, 2018-08-01

Re: [PATCH] perf tools: allow overriding MAX_NR_CPUS at compile time

From: Arnaldo Carvalho de Melo <acme@kernel.org>
Date: 2017-09-22 14:09:35
Also in: lkml

Em Fri, Sep 22, 2017 at 01:20:43PM +0200, Christophe Leroy escreveu:
quoted hunk ↗ jump to hunk
After update of kernel, perf tool doesn't run anymore on my
32MB RAM powerpc board, but still runs on a 128MB RAM board:

~# strace perf
execve("/usr/sbin/perf", ["perf"], [/* 12 vars */]) = -1 ENOMEM (Cannot allocate memory)
--- SIGSEGV {si_signo=SIGSEGV, si_code=SI_KERNEL, si_addr=0} ---
+++ killed by SIGSEGV +++
Segmentation fault

objdump -x shows that .bss section has a huge size of 24Mbytes:

 27 .bss          016baca8  101cebb8  101cebb8  001cd988  2**3

With especially the following objects having quite big size

10205f80 l     O .bss	00140000              runtime_cycles_stats
10345f80 l     O .bss	00140000              runtime_stalled_cycles_front_stats
10485f80 l     O .bss	00140000              runtime_stalled_cycles_back_stats
105c5f80 l     O .bss	00140000              runtime_branches_stats
10705f80 l     O .bss	00140000              runtime_cacherefs_stats
10845f80 l     O .bss	00140000              runtime_l1_dcache_stats
10985f80 l     O .bss	00140000              runtime_l1_icache_stats
10ac5f80 l     O .bss	00140000              runtime_ll_cache_stats
10c05f80 l     O .bss	00140000              runtime_itlb_cache_stats
10d45f80 l     O .bss	00140000              runtime_dtlb_cache_stats
10e85f80 l     O .bss	00140000              runtime_cycles_in_tx_stats
10fc5f80 l     O .bss	00140000              runtime_transaction_stats
11105f80 l     O .bss	00140000              runtime_elision_stats
11245f80 l     O .bss	00140000              runtime_topdown_total_slots
11385f80 l     O .bss	00140000              runtime_topdown_slots_retired
114c5f80 l     O .bss	00140000              runtime_topdown_slots_issued
11605f80 l     O .bss	00140000              runtime_topdown_fetch_bubbles
11745f80 l     O .bss	00140000              runtime_topdown_recovery_bubbles

This is due to commit 4d255766d28b1 ("perf: Bump max number of cpus
to 1024"), because many tables are sized with MAX_NR_CPUS

This patch gives the opportunity to redefine MAX_NR_CPUS via

make EXTRA_CFLAGS=-DMAX_NR_CPUS=1
I'll probably apply this, but the right thing would be to get rid of
MAX_NR_CPUS completely and instead allocate that using sysconf:

[root@jouet ~]# perf trace -e open,getdents getconf _NPROCESSORS_CONF
4
     0.014 ( 0.011 ms): getconf/2452 open(filename: /etc/ld.so.cache, flags: CLOEXEC                       ) = 3
     0.045 ( 0.008 ms): getconf/2452 open(filename: /lib64/libc.so.6, flags: CLOEXEC                       ) = 3
     0.264 ( 0.012 ms): getconf/2452 open(filename: /usr/lib/locale/locale-archive, flags: CLOEXEC         ) = 3
     0.315 ( 0.028 ms): getconf/2452 open(filename: /sys/devices/system/cpu, flags: CLOEXEC|DIRECTORY|NONBLOCK) = 3
     0.349 ( 0.013 ms): getconf/2452 getdents(fd: 3</sys/devices/system/cpu>, dirent: 0x944030, count: 32768) = 624
     0.365 ( 0.001 ms): getconf/2452 getdents(fd: 3</sys/devices/system/cpu>, dirent: 0x944030, count: 32768) = 0
[root@jouet ~]#
 
quoted hunk ↗ jump to hunk
Signed-off-by: Christophe Leroy <redacted>
---
 tools/perf/perf.h | 2 ++
 1 file changed, 2 insertions(+)
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index dc442ba21bf6..a9db563da0a9 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -23,7 +23,9 @@ static inline unsigned long long rdclock(void)
 	return ts.tv_sec * 1000000000ULL + ts.tv_nsec;
 }
 
+#ifndef MAX_NR_CPUS
 #define MAX_NR_CPUS			1024
+#endif
 
 extern const char *input_name;
 extern bool perf_host, perf_guest;
-- 
2.13.3
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help