Re: [PATCH 03/12] drm/msm/dsi: Add DSI PLL for 28nm 8960 PHY

From: Stephen Boyd <sboyd@codeaurora.org>
To: Archit Taneja <architt@codeaurora.org>
Cc: dri-devel@lists.freedesktop.org, robdclark@gmail.com,
	hali@codeaurora.org, linux-arm-msm@vger.kernel.org
Subject: Re: [PATCH 03/12] drm/msm/dsi: Add DSI PLL for 28nm 8960 PHY
Date: Fri, 16 Oct 2015 11:28:21 -0700	[thread overview]
Message-ID: <20151016182821.GC23912@codeaurora.org> (raw)
In-Reply-To: <5620F6D9.6050305@codeaurora.org>

On 10/16, Archit Taneja wrote:
> 
> 
> On 10/15/2015 02:05 AM, Stephen Boyd wrote:
> >On 10/14, Archit Taneja wrote:
> >>+	bytediv->hw.init = &bytediv_init;
> >>+	bytediv->reg = pll_28nm->mmio + REG_DSI_28nm_8960_PHY_PLL_CTRL_9;
> >>+
> >>+	snprintf(parent, 32, "dsi%dvco_clk", pll_28nm->id);
> >>+	snprintf(clk_name, 32, "dsi%dpllbyte", pll_28nm->id);
> >>+
> >>+	bytediv_init.name = clk_name;
> >>+	bytediv_init.ops = &clk_bytediv_ops;
> >>+	bytediv_init.flags = CLK_SET_RATE_PARENT;
> >>+	bytediv_init.parent_names = (const char *[]) { parent };
> >
> >Can't we just do &parent instead of this anonymous array?
> 
> &parent doesn't make sense here. parent in this function is an array
> of characters, not a pointer to a character.
> 
> I can think of only this way. We do something similar when we call
> clk_register_mux() in dsi_pll_28nm.c.
> 

Oh I missed that. Why not allocate the strings instead of putting
them on the stack?

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project