linux-kernel - [PATCH] drm/nv40/pm: silence GCC warnings

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>] [day] [month] [year] [list]

Message-ID: <1349730102.1397.24.camel@x61.thuisdomein>
Date:	Mon, 08 Oct 2012 23:01:42 +0200
From:	Paul Bolle <pebolle@...cali.nl>
To:	David Airlie <airlied@...ux.ie>
Cc:	dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org
Subject: [PATCH] drm/nv40/pm: silence GCC warnings

Building nv40_pm.o triggers these GCC warnings:
    drivers/gpu/drm/nouveau/nv40_pm.c: In function 'nv40_pm_clocks_pre':
    drivers/gpu/drm/nouveau/nv40_pm.c:164:41: warning: 'log2P' may be used uninitialized in this function [-Wmaybe-uninitialized]
    drivers/gpu/drm/nouveau/nv40_pm.c:165:38: warning: 'M2' may be used uninitialized in this function [-Wmaybe-uninitialized]
    drivers/gpu/drm/nouveau/nv40_pm.c:165:45: warning: 'M1' may be used uninitialized in this function [-Wmaybe-uninitialized]
    drivers/gpu/drm/nouveau/nv40_pm.c:165:25: warning: 'N2' may be used uninitialized in this function [-Wmaybe-uninitialized]
    drivers/gpu/drm/nouveau/nv40_pm.c:165:51: warning: 'N1' may be used uninitialized in this function [-Wmaybe-uninitialized]

But these variables seem to be initialized when used there. If the
preceding call of nv40_calc_pll() fails it will return a negative value
and this code will never be run. And if that call succeeds it will
return zero and all those five variables ought to be initialized when
used there.

Anyhow, it turns out that these warnings can be silenced by giving GCC
slightly more detailed information a little earlier. See,
get_pll_limits() returns an error-code integer (ie, negative on failure,
zero on success). And a trivial tweak to nv40_calc_pll() that takes this
into account makes these errors go away.

Signed-off-by: Paul Bolle <pebolle@...cali.nl>
---
0) I noticed these warnings while building recent releases (eg,
v3.6-rc7, v3.6, and v3.6.1) on current Fedora 17, using Fedora's default
config. But I found identical warnings in a build log for release
v3.4-rc1. I didn't bother checking earlier releases.

1) Compile tested only (I do not have the hardware involved at hand).

2) This is not the only place where get_pll_limits()'s return value is
checked. But this is apparently the only place where GCC needs to know
that any non-zero return will be negative.

 drivers/gpu/drm/nouveau/nv40_pm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nv40_pm.c b/drivers/gpu/drm/nouveau/nv40_pm.c
index e66273a..3df30cb 100644
--- a/drivers/gpu/drm/nouveau/nv40_pm.c
+++ b/drivers/gpu/drm/nouveau/nv40_pm.c
@@ -114,7 +114,7 @@ nv40_calc_pll(struct drm_device *dev, u32 reg, struct pll_lims *pll,
 	int ret;
 
 	ret = get_pll_limits(dev, reg, pll);
-	if (ret)
+	if (ret < 0)
 		return ret;
 
 	if (clk < pll->vco1.maxfreq)
-- 
1.7.11.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/