All of lore.kernel.org
 help / color / mirror / Atom feed
From: Rasmus Villemoes <linux@rasmusvillemoes.dk>
To: Kent Overstreet <kent.overstreet@gmail.com>,
	linux-kernel@vger.kernel.org
Cc: pmladek@suse.com, rostedt@goodmis.org
Subject: Re: [PATCH v3 03/33] vsprintf: Convert to printbuf
Date: Fri, 17 Jun 2022 10:59:02 +0200	[thread overview]
Message-ID: <0b4a9ba5-ed0d-e1a5-4982-8bad3e93f9b3@rasmusvillemoes.dk> (raw)
In-Reply-To: <ba8d21b8-7aae-6bfb-3553-3baba2bbeee1@gmail.com>

On 15/06/2022 20.44, Kent Overstreet wrote:
> On 6/15/22 05:09, Rasmus Villemoes wrote:

>> Anyway, my main concern with this is that performance goes down the
>> drain and the generated code will be awful. Have you done any
>> measurements and/or looked at disassembly? Thanks to
>> -fno-strict-aliasing (or perhaps just because we're writing through a
>> char* pointer which IIRC may alias anything), I think the compiler will
>> be forced to reload prt->pos and prt->size over and over and over. I may
>> be wrong, of course, that happens often. Perhaps __restrict could
>> help, IDK.
> 
> If we care that much about sprintf performance we must have some
> benchmarks somewhere - could you point me at them?

Try 'perf top'. Or copy the meat of vsprintf.c (all the %p gunk can be
elided) and do some measurements in userspace, that's what I did when I
improved the decimal conversion a few years back.

And yes, "we" care about sprintf performance because it's used to
generate all those files in /proc, /sys etc. etc. which many tools (e.g.
top/ps) rely on.

Rasmus

  reply	other threads:[~2022-06-17  8:59 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-04 19:30 [PATCH v3 00/33] Printbufs Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 01/33] lib/printbuf: New data structure for printing strings Kent Overstreet
2022-06-09 13:55   ` Petr Mladek
2022-06-04 19:30 ` [PATCH v3 02/33] lib/string_helpers: Convert string_escape_mem() to printbuf Kent Overstreet
2022-06-09 14:25   ` Petr Mladek
2022-06-09 17:53     ` Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 03/33] vsprintf: Convert " Kent Overstreet
2022-06-15  9:09   ` Rasmus Villemoes
2022-06-15 18:44     ` Kent Overstreet
2022-06-17  8:59       ` Rasmus Villemoes [this message]
2022-06-04 19:30 ` [PATCH v3 04/33] lib/hexdump: " Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 05/33] vsprintf: %pf(%p) Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 06/33] lib/string_helpers: string_get_size() now returns characters wrote Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 07/33] lib/printbuf: Heap allocation Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 08/33] lib/printbuf: Tabstops, indenting Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 09/33] lib/printbuf: Unit specifiers Kent Overstreet
2022-06-05  1:06   ` Joe Perches
2022-06-04 19:30 ` [PATCH v3 10/33] lib/pretty-printers: prt_string_option(), prt_bitflags() Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 11/33] vsprintf: Improve number() Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 12/33] vsprintf: prt_u64_minwidth(), prt_u64() Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 13/33] test_printf: Drop requirement that sprintf not write past nul Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 14/33] vsprintf: Start consolidating printf_spec handling Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 15/33] vsprintf: Refactor resource_string() Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 16/33] vsprintf: Refactor fourcc_string() Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 17/33] vsprintf: Refactor ip_addr_string() Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 18/33] vsprintf: Refactor mac_address_string() Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 19/33] vsprintf: time_and_date() no longer takes printf_spec Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 20/33] vsprintf: flags_string() " Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 21/33] vsprintf: Refactor device_node_string, fwnode_string Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 22/33] vsprintf: Refactor hex_string, bitmap_string_list, bitmap_string Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 23/33] Input/joystick/analog: Convert from seq_buf -> printbuf Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 24/33] mm/memcontrol.c: Convert to printbuf Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 25/33] clk: tegra: bpmp: " Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 26/33] tools/testing/nvdimm: " Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 27/33] powerpc: " Kent Overstreet
2022-06-04 19:30   ` Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 28/33] x86/resctrl: " Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 29/33] PCI/P2PDMA: " Kent Overstreet
2022-06-08 21:11   ` Bjorn Helgaas
2022-06-08 23:24     ` Kent Overstreet
2022-06-08 23:34       ` Logan Gunthorpe
2022-06-08 23:40       ` Bjorn Helgaas
2022-06-04 19:30 ` [PATCH v3 30/33] tracing: trace_events_synth: " Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 31/33] d_path: prt_path() Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 32/33] tracing: Convert to printbuf Kent Overstreet
2022-06-04 19:30 ` [PATCH v3 33/33] Delete seq_buf Kent Overstreet
2022-06-05 16:21 ` [PATCH v3 00/33] Printbufs Steven Rostedt
2022-06-05 17:55   ` Kent Overstreet

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=0b4a9ba5-ed0d-e1a5-4982-8bad3e93f9b3@rasmusvillemoes.dk \
    --to=linux@rasmusvillemoes.dk \
    --cc=kent.overstreet@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pmladek@suse.com \
    --cc=rostedt@goodmis.org \
    /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.