From: Ben Widawsky Subject: Re: [PATCH 5/7] drm/i915: WARN on unclaimed registers Date: Fri, 25 Jan 2013 17:11:24 -0800 Message-ID: <20130125171124.5b3e4cba@bwidawsk.net> References: <1359147462-3902-1-git-send-email-przanoni@gmail.com> <1359147462-3902-6-git-send-email-przanoni@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from shiva.chad-versace.us (209-20-75-48.static.cloud-ips.com [209.20.75.48]) by gabe.freedesktop.org (Postfix) with ESMTP id 625D9E5FB7 for ; Fri, 25 Jan 2013 17:10:08 -0800 (PST) In-Reply-To: <1359147462-3902-6-git-send-email-przanoni@gmail.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org Errors-To: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org To: Paulo Zanoni Cc: intel-gfx@lists.freedesktop.org, Paulo Zanoni List-Id: intel-gfx@lists.freedesktop.org On Fri, 25 Jan 2013 18:57:40 -0200 Paulo Zanoni wrote: > From: Paulo Zanoni > > While debugging these "unclaimed register" problems I concluded that > having a backtrace is way much more useful than having the register > address, since in a lot of cases the register address print on the > message is not the register we're looking for. > > We must fix all the "unclaimed register" problems, so if dmesg gets > too polluted it means we're too bugged. > > Signed-off-by: Paulo Zanoni As I mentioned internally, I'd still prefer if (WARN_ON(I915_READ_NOTRACE(FPGA_DBG) & FPGA_DBG_RM_NOCLAIM)) because I don't think the message is particularly useful with a backtrace in hand. Also on second thought since the internal review, it's probably not super useful to have the WARN in the UNCLAIMED_REG_CLEAR, because as you've mentioned before, it's usually from snd-hda or something like that. So I'd probably get rid of that unless someone else sees a super cool usage for it. With or without painting my color: Reviewed-by: Ben Widawsky > --- > drivers/gpu/drm/i915/i915_drv.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c > index 422dfc6..bc0eb88 100644 > --- a/drivers/gpu/drm/i915/i915_drv.c > +++ b/drivers/gpu/drm/i915/i915_drv.c > @@ -1227,14 +1227,14 @@ ilk_dummy_write(struct drm_i915_private *dev_priv) > #define UNCLAIMED_REG_CLEAR(dev_priv, reg, op) \ > if (IS_HASWELL(dev_priv->dev) && \ > (I915_READ_NOTRACE(FPGA_DBG) & FPGA_DBG_RM_NOCLAIM)) { \ > - DRM_ERROR("Unclaimed register before %x (%c)\n", reg, op); \ > + WARN(1, "Unclaimed register before %x (%c)\n", reg, op); \ > I915_WRITE_NOTRACE(FPGA_DBG, FPGA_DBG_RM_NOCLAIM); \ > } > > #define UNCLAIMED_REG_CHECK(dev_priv, reg, op) \ > if (IS_HASWELL(dev_priv->dev) && \ > (I915_READ_NOTRACE(FPGA_DBG) & FPGA_DBG_RM_NOCLAIM)) { \ > - DRM_ERROR("Unclaimed register %x (%c)\n", reg, op); \ > + WARN(1, "Unclaimed register %x (%c)\n", reg, op); \ > I915_WRITE_NOTRACE(FPGA_DBG, FPGA_DBG_RM_NOCLAIM); \ > } > -- Ben Widawsky, Intel Open Source Technology Center