From: Sam Ravnborg <sam@ravnborg.org>
To: Andi Kleen <andi@firstfloor.org>
Cc: linux-kernel@vger.kernel.org, akpm@linux-foundation.org,
Andi Kleen <ak@linux.intel.com>,
linux-kbuild@vger.kernel.org, mmarek@suse.cz
Subject: Re: [PATCH 8/8] Kbuild: add inline-account tool to find inline bloat
Date: Sat, 17 May 2014 11:36:15 +0200 [thread overview]
Message-ID: <20140517093615.GA16926@ravnborg.org> (raw)
In-Reply-To: <20140517083144.GA16691@ravnborg.org>
On Sat, May 17, 2014 at 10:31:44AM +0200, Sam Ravnborg wrote:
> Hi Andi.
>
> On Fri, May 16, 2014 at 02:43:15PM -0700, Andi Kleen wrote:
> > From: Andi Kleen <ak@linux.intel.com>
> >
> > Add a tool to hunt for inline bloat. It uses objdump -S to account
> > inlines.
> I tried this on my sparc32 build - but it failed with:
> objdump: can't disassemble for architecture UNKNOWN!
>
> It looks simple to add CROSS_COMPILE support but I did not do so.
> My python skills are non-existing.
Patched the calls to nm and objdump - but it gave no output
when I ran the script.
nm --print-size shows following output:
00002910 00000024 r CSWTCH.946
00002bd4 00000024 r CSWTCH.951
U PDE_DATA
U ROOT_DEV
000000fc 00000014 T SyS_accept
00002c98 000001a8 T SyS_accept4
00000fc4 0000008c T SyS_bind
00000eb4 00000094 T SyS_connect
00000d98 00000094 T SyS_getpeername
00000e2c 00000088 T SyS_getsockname
00000c6c 00000090 T SyS_getsockopt
00000f48 0000007c T SyS_listen
00000128 00000018 T SyS_recv
0000142c 000000f0 T SyS_recvfrom
00001920 000000d0 T SyS_recvmmsg
00001238 00000010 T SyS_recvmsg
00000110 00000018 T SyS_send
00001d38 00000010 T SyS_sendmmsg
00001db0 00000010 T SyS_sendmsg
000015f4 000000dc T SyS_sendto
00000cfc 0000009c T SyS_setsockopt
00000c0c 00000060 T SyS_shutdown
00003020 000000b4 T SyS_socket
000007ec 000001f8 T SyS_socketcall
00002e40 000001e0 T SyS_socketpair
000b776c 00000098 t T.1063
000b762c 000000d0 t T.1064
objdump -Sl shows following output:
000000d4 <sock_mmap>:
sock_mmap():
d4: 9d e3 bf a0 save %sp, -96, %sp
d8: c2 06 20 78 ld [ %i0 + 0x78 ], %g1
dc: 94 10 00 19 mov %i1, %o2
e0: 92 10 00 01 mov %g1, %o1
e4: c2 00 60 18 ld [ %g1 + 0x18 ], %g1
e8: c2 00 60 40 ld [ %g1 + 0x40 ], %g1
ec: 9f c0 40 00 call %g1
f0: 90 10 00 18 mov %i0, %o0
f4: 81 c7 e0 08 ret
f8: 91 e8 00 08 restore %g0, %o0, %o0
000000fc <SyS_accept>:
sys_accept():
fc: 96 10 20 00 clr %o3
100: 82 13 c0 00 mov %o7, %g1
104: 40 00 00 00 call 104 <SyS_accept+0x8>
108: 9e 10 40 00 mov %g1, %o7
10c: 01 00 00 00 nop
00000110 <SyS_send>:
SyS_send():
110: 98 10 20 00 clr %o4 ! 0 <sock_from_file-0x4c>
114: 9a 10 20 00 clr %o5
118: 82 13 c0 00 mov %o7, %g1
11c: 40 00 00 00 call 11c <SyS_send+0xc>
120: 9e 10 40 00 mov %g1, %o7
124: 01 00 00 00 nop
00000128 <SyS_recv>:
sys_recv():
128: 98 10 20 00 clr %o4 ! 0 <sock_from_file-0x4c>
12c: 9a 10 20 00 clr %o5
130: 82 13 c0 00 mov %o7, %g1
134: 40 00 00 00 call 134 <SyS_recv+0xc>
138: 9e 10 40 00 mov %g1, %o7
13c: 01 00 00 00 nop
Sam
next prev parent reply other threads:[~2014-05-17 9:36 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-16 21:43 Fix some common inline bloat Andi Kleen
2014-05-16 21:43 ` [PATCH 1/8] ixgbe: Out of line ixgbe_read/write_reg Andi Kleen
2014-05-19 9:14 ` David Laight
2014-05-19 22:00 ` Rustad, Mark D
2014-05-19 23:25 ` Andi Kleen
2014-05-20 17:06 ` Rustad, Mark D
2014-05-16 21:43 ` [PATCH 2/8] radeonfb: Out of line errata workarounds Andi Kleen
2014-05-16 21:43 ` Andi Kleen
2014-05-16 21:43 ` [PATCH 3/8] list: Out of line INIT_LIST_HEAD and list_del Andi Kleen
2014-05-17 0:03 ` Dave Jones
2014-05-17 2:37 ` Andi Kleen
2014-05-17 0:03 ` Eric Dumazet
2014-05-16 21:43 ` [PATCH 4/8] e1000e: Out of line __ew32_prepare/__ew32 Andi Kleen
2014-05-17 3:23 ` Stephen Hemminger
2014-05-16 21:43 ` [PATCH 5/8] x86: Out of line get_dma_ops Andi Kleen
2014-05-16 21:43 ` [PATCH 6/8] ftrace: Out of line ftrace_trigger_soft_disabled Andi Kleen
2014-05-16 21:43 ` [PATCH 7/8] radeon: Out of line radeon_get_ib_value Andi Kleen
2014-05-20 16:16 ` Marek Olšák
2014-05-20 16:16 ` Marek Olšák
2014-05-20 17:04 ` Andi Kleen
2014-05-20 18:14 ` Christian König
2014-05-16 21:43 ` [PATCH 8/8] Kbuild: add inline-account tool to find inline bloat Andi Kleen
2014-05-17 8:31 ` Sam Ravnborg
2014-05-17 9:36 ` Sam Ravnborg [this message]
2014-05-17 16:51 ` Andi Kleen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140517093615.GA16926@ravnborg.org \
--to=sam@ravnborg.org \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mmarek@suse.cz \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.