From mboxrd@z Thu Jan 1 00:00:00 1970 X-Msuck: nntp://news.gmane.org/gmane.linux.lib.musl.general/7365 Path: news.gmane.org!not-for-mail From: John Mudd Newsgroups: gmane.linux.lib.musl.general Subject: Re: musl perf, 20% slower than native build? Date: Wed, 8 Apr 2015 17:48:37 -0400 Message-ID: References: <20150408160507.GB31681@port70.net> Reply-To: musl@lists.openwall.com NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: multipart/alternative; boundary=20cf3022404b94453005133d7f0b X-Trace: ger.gmane.org 1428529766 15148 80.91.229.3 (8 Apr 2015 21:49:26 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Wed, 8 Apr 2015 21:49:26 +0000 (UTC) Cc: musl , John Mudd To: Alexander Monakov Original-X-From: musl-return-7378-gllmg-musl=m.gmane.org@lists.openwall.com Wed Apr 08 23:49:17 2015 Return-path: Envelope-to: gllmg-musl@m.gmane.org Original-Received: from mother.openwall.net ([195.42.179.200]) by plane.gmane.org with smtp (Exim 4.69) (envelope-from ) id 1YfxqS-0004z2-Dz for gllmg-musl@m.gmane.org; Wed, 08 Apr 2015 23:49:12 +0200 Original-Received: (qmail 16003 invoked by uid 550); 8 Apr 2015 21:49:10 -0000 Mailing-List: contact musl-help@lists.openwall.com; run by ezmlm Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: Original-Received: (qmail 15984 invoked from network); 8 Apr 2015 21:49:09 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=RHNiAjAvptiFYj0VdK61Es2HQKmCNbXkOqQe/2ARW0I=; b=KeHqyw75dC9bypbieZCWtY9Bk09nFy6bS2pvAP47lIwSrn+IBB/UCx6VqqUfS+nSmM 98IS5jth9wlFfrpDCC0x3SvdiqbALmQpQHqP72wgH4KYRoyg8Y6+QuFNQ6MI6b7vrqeY pEvDjN4ZHIRGGwtGb8M3iFsELN4YQMu2JGQ3Xa4ztJiN8XoMEX+VbIc0Ol/V5LnLagq3 X1tS+kkcp503cqNtW9rq6dc+hov5hPwnM0RL9fdSKEXt9qqaDllTXVDp+AQ+moJyisPC +0vm29nluH5Gtq5U4nnyXNPhzxNpMOoWXmEVtb1HeO2GymoSmxUVKaHyJ44vG1Z8O47k b6JQ== X-Received: by 10.42.86.12 with SMTP id s12mr35855261icl.47.1428529737907; Wed, 08 Apr 2015 14:48:57 -0700 (PDT) In-Reply-To: Xref: news.gmane.org gmane.linux.lib.musl.general:7365 Archived-At: --20cf3022404b94453005133d7f0b Content-Type: text/plain; charset=KOI8-R Reran benchmark using -O3 musl library and now musl and native have basically the same run times. native: 2.08% python libc-2.19.so [.] __memcpy_ssse3 0.81% python libc-2.19.so [.] __x86.get_pc_thunk.bx 0.56% python libc-2.19.so [.] __memset_sse2 0.53% python libc-2.19.so [.] _int_malloc 0.45% python libc-2.19.so [.] _int_free 0.40% python libc-2.19.so [.] malloc 0.22% python libc-2.19.so [.] realloc 0.21% python libc-2.19.so [.] __ctype_b_loc 0.12% python libc-2.19.so [.] __x86.get_pc_thunk.cx 0.12% python libc-2.19.so [.] free 0.01% python libc-2.19.so [.] __strlen_sse2_bsf 0.01% python libc-2.19.so [.] malloc_consolidate 0.01% python libc-2.19.so [.] __memcpy_ia32 0.00% python libc-2.19.so [.] _IO_getc 0.00% python libc-2.19.so [.] mremap_chunk 0.00% python libc-2.19.so [.] _int_realloc 0.00% python libc-2.19.so [.] __strcmp_ssse3 musl: 5.37% python libc.so [.] memcpy 1.96% python libc.so [.] malloc 1.93% python libc.so [.] free 0.60% python libc.so [.] memset 0.20% python libc.so [.] __x86.get_pc_thunk.bx 0.12% python libc.so [.] realloc 0.03% python libc.so [.] alloc_fwd 0.02% python libc.so [.] alloc_rev 0.01% python libc.so [.] find_sym 0.01% python libc.so [.] strcmp 0.00% python libc.so [.] memcmp 0.00% python libc.so [.] __vm_lock 0.00% python libc.so [.] __strerror_l 0.00% python libc.so [.] __cp_end 0.00% python libc.so [.] .L350 0.00% python libc.so [.] strlen 0.00% python libc.so [.] memchr 0.00% python libc.so [.] pthread_rwlock_trywrlock On Wed, Apr 8, 2015 at 4:24 PM, John Mudd wrote: > Ignore my previous email. The symbols are there. I'll rebuild again and > check the speed... > > On Wed, Apr 8, 2015 at 4:16 PM, John Mudd wrote: > >> I don't see any of those "a_*" names in the output of 'nm' or 'readelf >> -a' on musl-1.1.8/lib/libc.so. >> >>> >>> >> > --20cf3022404b94453005133d7f0b Content-Type: text/html; charset=KOI8-R Content-Transfer-Encoding: quoted-printable
Reran benchmark using -O3 musl library and now musl and na= tive have basically the same run times.

native:
=9A =9A =9A2.08% =9A python =9A= libc-2.19.so =9A =9A =9A =9A [.] __memcpy_ssse3
=9A =9A =9A0.= 81% =9A python =9Alibc-2.19.so =9A =9A = =9A =9A [.] __x86.get_pc_thunk.bx
=9A =9A =9A0.56% =9A python =9A= libc-2.19.so =9A =9A =9A =9A [.] __mems= et_sse2
=9A =9A =9A0.53% =9A python =9Alibc-2.19.so =9A =9A =9A =9A [.] _int_malloc
=9A =9A = =9A0.45% =9A python =9Alibc-2.19.so =9A= =9A =9A =9A [.] _int_free
=9A =9A =9A0.40% =9A python =9Alibc-2.19.so =9A =9A =9A =9A [.] malloc
<= div>=9A =9A =9A0.22% =9A python =9Alibc-2.1= 9.so =9A =9A =9A =9A [.] realloc
=9A =9A =9A0.21% =9A python = =9Alibc-2.19.so =9A =9A =9A =9A [.] __c= type_b_loc
=9A =9A =9A0.12% =9A python =9Alibc-2.19.so =9A =9A =9A =9A [.] __x86.get_pc_thunk.cx
=9A =9A =9A0.12% =9A python = =9Alibc-2.19.so =9A =9A =9A =9A [.] fre= e
=9A =9A =9A0.01% =9A python =9A= libc-2.19.so =9A =9A =9A =9A [.] __strlen_sse2_bsf
=9A =9A = =9A0.01% =9A python =9Alibc-2.19.so =9A= =9A =9A =9A [.] malloc_consolidate
=9A =9A =9A0.01% =9A python = =9Alibc-2.19.so =9A =9A =9A =9A [.] __m= emcpy_ia32
=9A =9A =9A0.00% =9A python =9Alibc-2.19.so =9A =9A =9A =9A [.] _IO_getc
=9A =9A = =9A0.00% =9A python =9Alibc-2.19.so =9A= =9A =9A =9A [.] mremap_chunk
=9A =9A =9A0.00% =9A python =9Alibc-2.19.so =9A =9A =9A =9A [.] _int_reall= oc
=9A =9A =9A0.00% =9A python =9Alibc-2.19.so =9A =9A =9A =9A [.] __strcmp_ssse3
=

musl:
=9A =9A =9A5.37% =9A pyt= hon =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] memcpy
=9A =9A =9A1= .96% =9A python =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] malloc
= =9A =9A =9A1.93% =9A python =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] free<= /div>
=9A =9A =9A0.60% =9A python =9Alibc.so =9A =9A =9A =9A =9A =9A = =9A[.] memset
=9A =9A =9A0.20% =9A python =9Alibc.so =9A =9A =9A = =9A =9A =9A =9A[.] __x86.get_pc_thunk.bx
=9A =9A =9A0.12% =9A pyt= hon =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] realloc
=9A =9A =9A= 0.03% =9A python =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] alloc_fwd
<= div>=9A =9A =9A0.02% =9A python =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] a= lloc_rev
=9A =9A =9A0.01% =9A python =9Alibc.so =9A =9A =9A =9A = =9A =9A =9A[.] find_sym
=9A =9A =9A0.01% =9A python =9Alibc.so = =9A =9A =9A =9A =9A =9A =9A[.] strcmp
=9A =9A =9A0.00% =9A python= =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] memcmp
=9A =9A =9A0.00= % =9A python =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] __vm_lock
= =9A =9A =9A0.00% =9A python =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] __str= error_l
=9A =9A =9A0.00% =9A python =9Alibc.so =9A =9A =9A =9A = =9A =9A =9A[.] __cp_end
=9A =9A =9A0.00% =9A python =9Alibc.so = =9A =9A =9A =9A =9A =9A =9A[.] .L350
=9A =9A =9A0.00% =9A python = =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] strlen
=9A =9A =9A0.00%= =9A python =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] memchr
=9A = =9A =9A0.00% =9A python =9Alibc.so =9A =9A =9A =9A =9A =9A =9A[.] pthread_r= wlock_trywrlock




On Wed, Apr 8, 2015 = at 4:24 PM, John Mudd <johnbmudd@gmail.com> wrote:
Ignore my previous email. The symbols are the= re. I'll rebuild again and check the speed...

On Wed, Apr 8, 2015 at 4:16 PM, J= ohn Mudd <johnbmudd@gmail.com> wrote:
=
I don't see any of those "a_*"= ; names in the output of 'nm' or 'readelf -a' on musl-1.1.8= /lib/libc.so.




--20cf3022404b94453005133d7f0b--