On Tue, Dec 4, 2012 at 6:51 PM, Juri Lelli juri.lelli@gmail.com wrote:
Hi all, in the process of testing scheduler patches on a Pandaboard ES, I found a possible regression regarding OMAP mmc. The system is a Pandaboard ES (OMAP4460 ES1.1) running Linaro 12.10 (GNU/Linux 3.4.0-2-linaro-lt-omap armv7l).
The problem is that the system remains stuck at boot time trying to mount the root device.
This is what I got via git bisect from Linux 3.5 and Linux 3.7-rc7:
26b88520b80695a6fa5fd95b5d97c03f4daf87e0 is the first bad commit commit 26b88520b80695a6fa5fd95b5d97c03f4daf87e0 Author: Russell King rmk+kernel@arm.linux.org.uk Date: Fri Apr 13 12:27:37 2012 +0100
mmc: omap_hsmmc: remove private DMA API implementation Remove the private DMA API implementation from omap_hsmmc, making it use entirely the DMA engine API. Tested-by: Tony Lindgren <tony@atomide.com> Tested-by: Venkatraman S <svenkatr@ti.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>:040000 040000 03cff4d68cc92ea652ef8e0c6ac74af804bc9f00 a9763c6b364bfc66b5fcd4cada7315befcbd62e8 M drivers
Here instead the two boot traces, before (all OK) and after the bad commit.
[ 1.661041] sdhci: Secure Digital Host Controller Interface driver [ 1.667602] sdhci: Copyright(c) Pierre Ossman [ 1.673736] omap_hsmmc omap_hsmmc.0: Failed to get debounce clk [ 1.679992] omap_hsmmc omap_hsmmc.0: unable to obtain RX DMA engine channel 62 [ 1.687774] omap_hsmmc omap_hsmmc.0: unable to obtain TX DMA engine channel 61 [ 1.745422] omap_hsmmc omap_hsmmc.4: Failed to get debounce clk [ 1.751647] omap_hsmmc omap_hsmmc.4: unable to obtain RX DMA engine channel 60 [ 1.759277] omap_hsmmc omap_hsmmc.4: unable to obtain TX DMA engine channel 59
Change:
# CONFIG_DMADEVICES is not set
To: CONFIG_DMADEVICES=y # CONFIG_DMADEVICES_DEBUG is not set
# # DMA Devices # # CONFIG_DW_DMAC is not set CONFIG_TIMB_DMA=m # CONFIG_TI_EDMA is not set CONFIG_DMA_OMAP=y CONFIG_DMA_ENGINE=y CONFIG_DMA_VIRTUAL_CHANNELS=y
Regards,