On 16/07/2013 20:47, Alan Stern wrote:
On Tue, 16 Jul 2013, boris brezillon wrote:+ uclk = clk_get(&pdev->dev, "usb_clk"); + if (IS_ERR(uclk)) { + uclk = NULL; + dev_warn(&pdev->dev, "failed to get usb_clk\n"); + }Is this really what you want for backward compatibility?Here are some proposition to remove the warning message: 1) replace it with a dev_info and change the message: dev_info(&pdev->dev, "failed to get usb_clk (most likely using old at91 clk implementation)\n"); 2) drop the log and silently ignore the missing clk (I'm not a big fan of this solution as it may lead to some errors if we're using new clk implem and the clock is really missing) 3) rework the current clk_set_rate function to accept clk_set_rate on usb clk and add clk_lookup entries for the usb clk (I'm not a big fan of this solution neither as this modifications will only be used for a short time until the transition to common clk framework is completed).Another possibility is to combine this change with the clock implementation update, and do them in a single patch. Then backward compatibility would not be an issue.
Yes, that was one of the question I asked in the cover-letter. I think I'll move these patches in the "move to common clk" series. Thanks
Alan Stern
-- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
