lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date:   Sat, 10 Jul 2021 23:49:58 -0600
From:   Jim Cromie <jim.cromie@...il.com>
To:     Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
        Maxime Ripard <mripard@...nel.org>,
        Thomas Zimmermann <tzimmermann@...e.de>,
        David Airlie <airlied@...ux.ie>,
        Daniel Vetter <daniel@...ll.ch>,
        Zhenyu Wang <zhenyuw@...ux.intel.com>,
        Zhi Wang <zhi.a.wang@...el.com>,
        Jani Nikula <jani.nikula@...ux.intel.com>,
        Joonas Lahtinen <joonas.lahtinen@...ux.intel.com>,
        Rodrigo Vivi <rodrigo.vivi@...el.com>,
        dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org,
        intel-gvt-dev@...ts.freedesktop.org,
        intel-gfx@...ts.freedesktop.org
Cc:     jbaron@...mai.com, Jim Cromie <jim.cromie@...il.com>
Subject: [RFC PATCH v2 0/4] Allow using dyndbg to replace drm_debug_enabled

drm_debug_enabled() is called a lot to do unlikely bit-tests to
control debug printing; this is a good job for dynamic-debug, IFF it
is built with JUMP_LABEL.
 
Enable the use of dynamic-debug to avoid drm_debug_enabled()
overheads, opt in with CONFIG_DRM_USE_DYNAMIC_DEBUG=y.

I have this patchset running bare-metal on an i915 laptop & an amdgpu
desktop (both as loadable modules).

I booted the amdgpu box with:

BOOT_IMAGE=(hd2,gpt2)/vmlinuz-5.13.0-dd7-13692-g8def25788f56 \
     root=UUID=mumble ro \
     rootflags=subvol=root00 rhgb \
     dynamic_debug.verbose=3 main.dyndbg=+p \
     amdgpu.debug=1 amdgpu.test=1 \
     "amdgpu.dyndbg=format ^[ +p"

That last line activates ~1700 callsites with a format like '[DML' etc
at boot, causing ~76k prdbgs in 409 seconds, before I turned them off
with:

  echo module amdgpu -p > /proc/dynamic_debug/control

[root@...dalf jimc]# journalctl -b-0 | grep -P '\[(DML|VBLANK|SURFACE|BIOS|BANDWIDTH)' | wc
  68708  578503 5054437
[root@...dalf jimc]# journalctl -b-0 | grep -P '\[(DML|VBLANK|SURFACE|BIOS|BANDWIDTH|\w+)' | wc
  76298  661176 6028087

IOW, things appear to hold up under some stress.

this is on top of master @ v5.13-13688-gde5540965853

v1 is here:
https://lore.kernel.org/lkml/20201204035318.332419-1-jim.cromie@gmail.com/

Jim Cromie (4):
  drm_print.h: rewrap __DRM_DEFINE_DBG_RATELIMITED macro
  drm: fixup comment spelling
  drm: RFC add choice to use dynamic debug in drm-debug
  i915: map gvt pr_debug categories to bits in parameters/debug_gvt

 drivers/gpu/drm/Kconfig            |  13 ++++
 drivers/gpu/drm/drm_print.c        |  75 +++++++++++++++++-
 drivers/gpu/drm/i915/gvt/Makefile  |   4 +
 drivers/gpu/drm/i915/i915_params.c |  76 ++++++++++++++++++
 include/drm/drm_print.h            | 119 ++++++++++++++++++++---------
 5 files changed, 249 insertions(+), 38 deletions(-)

-- 
2.31.1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ