Thread (16 messages) 16 messages, 4 authors, 2012-05-16

[PATCH 3/4] ARM: tegra30: Add Tegra Memory Controller(MC) driver

From: Hiroshi Doyu <hidden>
Date: 2012-05-11 06:17:12
Also in: linux-devicetree, linux-tegra, lkml

Hi Greg,

From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Subject: Re: [PATCH 3/4] ARM: tegra30: Add Tegra Memory Controller(MC) driver
Date: Thu, 10 May 2012 23:46:04 +0200
Message-ID: [ref]
On Thu, May 10, 2012 at 10:42:32AM +0300, Hiroshi DOYU wrote:
quoted
Tegra Memory Controller(MC) driver for Tegra30
Added to support MC General interrupts, mainly for IOMMU(SMMU).
I'll apply this, but:
Thanks.
quoted
+static inline u32 mc_readl(struct tegra30_mc *mc, u32 offs)
+{
+	if (offs < 0x10)
+		return readl(mc->regs[0] + offs);
+	BUG_ON(offs < 0x3c);
+	if (offs < 0x1f0)
+		return readl(mc->regs[1] + offs - 0x3c);
+	BUG_ON(offs < 0x200);
+	if (offs < 0x228)
+		return readl(mc->regs[2] + offs - 0x200);
+	BUG_ON(offs < 0x284);
+	if (offs < 0x400)
+		return readl(mc->regs[3] + offs - 0x284);
+	BUG();
+}
That's a lot of BUG* calls.  Same thing with the 1/4 patch in this
series.

We really should not have a BUG call in any driver, as you just stopped
the whole system.  I can understand this being there for debugging when
you create the code originally, but as these are things no one should
ever be able to hit now, you should remove them, right?

Care to write a follow-on patch removing all of these BUG_ON and BUG
calls in this and the drivers/memory/tegra20-mc.c driver?
Ok, they are replying to this email.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help