Message ID | 1421125757-10794-1-git-send-email-b38343@freescale.com (mailing list archive) |
---|---|
State | Superseded |
Headers | show |
On Tue, Jan 13, 2015 at 01:09:17PM +0800, Robin Gong wrote: > Below comments got from Page4724 of Reference Manual of i.mx6q: > http://cache.freescale.com/files/32bit/doc/ref_manual/IMX6DQRM.pdf > > --"Static context mode should be used for the first channel called > after reset to ensure that the all context RAM for that channel is > initialized during the context SAVE phase when the channel is > done or yields. Subsequent calls to the same channel or > different channels may use any of the dynamic context modes. > This will ensure that all context locations for the bootload > channel are initialized, and prevent undefined values in context > RAM from being loaded during the context restore if the > channel is re-started later" > > Unfortunately, the rule was broken by commit(5b28aa319bba96987316425a1131813d87cbab35) > .This patch just take them back. > > Signed-off-by: Robin Gong <b38343@freescale.com> > --- > drivers/dma/imx-sdma.c | 7 ++++--- > 1 file changed, 4 insertions(+), 3 deletions(-) > > diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c > index d0df198..510e463 100644 > --- a/drivers/dma/imx-sdma.c > +++ b/drivers/dma/imx-sdma.c > @@ -1323,6 +1323,10 @@ static void sdma_load_firmware(const struct firmware *fw, void *context) > sdma_load_script(sdma, ram_code, > header->ram_code_size, > addr->ram_code_start_addr); > + > + /* Set bits of CONFIG register with given context switching mode */ > + writel_relaxed(SDMA_H_CONFIG_CSM, sdma->regs + SDMA_H_CONFIG); > + You move the CONFIG register initialization to a place which may never be called. Look at the top of sdma_load_firmware: if (!fw) { dev_info(sdma->dev, "external firmware not found, using ROM firmware\n"); /* In this case we just use the ROM firmware. */ return; } Sascha
On Tue, Jan 20, 2015 at 08:16:14AM +0100, Sascha Hauer wrote: > On Tue, Jan 13, 2015 at 01:09:17PM +0800, Robin Gong wrote: > > Below comments got from Page4724 of Reference Manual of i.mx6q: > > http://cache.freescale.com/files/32bit/doc/ref_manual/IMX6DQRM.pdf > > > > --"Static context mode should be used for the first channel called > > after reset to ensure that the all context RAM for that channel is > > initialized during the context SAVE phase when the channel is > > done or yields. Subsequent calls to the same channel or > > different channels may use any of the dynamic context modes. > > This will ensure that all context locations for the bootload > > channel are initialized, and prevent undefined values in context > > RAM from being loaded during the context restore if the > > channel is re-started later" > > > > Unfortunately, the rule was broken by commit(5b28aa319bba96987316425a1131813d87cbab35) > > .This patch just take them back. > > > > Signed-off-by: Robin Gong <b38343@freescale.com> > > --- > > drivers/dma/imx-sdma.c | 7 ++++--- > > 1 file changed, 4 insertions(+), 3 deletions(-) > > > > diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c > > index d0df198..510e463 100644 > > --- a/drivers/dma/imx-sdma.c > > +++ b/drivers/dma/imx-sdma.c > > @@ -1323,6 +1323,10 @@ static void sdma_load_firmware(const struct firmware *fw, void *context) > > sdma_load_script(sdma, ram_code, > > header->ram_code_size, > > addr->ram_code_start_addr); > > + > > + /* Set bits of CONFIG register with given context switching mode */ > > + writel_relaxed(SDMA_H_CONFIG_CSM, sdma->regs + SDMA_H_CONFIG); > > + > > You move the CONFIG register initialization to a place which may never > be called. Look at the top of sdma_load_firmware: > > if (!fw) { > dev_info(sdma->dev, "external firmware not found, using ROM firmware\n"); > /* In this case we just use the ROM firmware. */ > return; > } > > Sascha > Yes. I'll move it into sdma_run_channel0. > -- > Pengutronix e.K. | | > Industrial Linux Solutions | http://www.pengutronix.de/ | > Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | > Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | -- To unsubscribe from this list: send the line "unsubscribe dmaengine" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c index d0df198..510e463 100644 --- a/drivers/dma/imx-sdma.c +++ b/drivers/dma/imx-sdma.c @@ -1323,6 +1323,10 @@ static void sdma_load_firmware(const struct firmware *fw, void *context) sdma_load_script(sdma, ram_code, header->ram_code_size, addr->ram_code_start_addr); + + /* Set bits of CONFIG register with given context switching mode */ + writel_relaxed(SDMA_H_CONFIG_CSM, sdma->regs + SDMA_H_CONFIG); + clk_disable(sdma->clk_ipg); clk_disable(sdma->clk_ahb); @@ -1401,9 +1405,6 @@ static int sdma_init(struct sdma_engine *sdma) writel_relaxed(ccb_phys, sdma->regs + SDMA_H_C0PTR); - /* Set bits of CONFIG register with given context switching mode */ - writel_relaxed(SDMA_H_CONFIG_CSM, sdma->regs + SDMA_H_CONFIG); - /* Initializes channel's priorities */ sdma_set_channel_priority(&sdma->channel[0], 7);
Below comments got from Page4724 of Reference Manual of i.mx6q: http://cache.freescale.com/files/32bit/doc/ref_manual/IMX6DQRM.pdf --"Static context mode should be used for the first channel called after reset to ensure that the all context RAM for that channel is initialized during the context SAVE phase when the channel is done or yields. Subsequent calls to the same channel or different channels may use any of the dynamic context modes. This will ensure that all context locations for the bootload channel are initialized, and prevent undefined values in context RAM from being loaded during the context restore if the channel is re-started later" Unfortunately, the rule was broken by commit(5b28aa319bba96987316425a1131813d87cbab35) .This patch just take them back. Signed-off-by: Robin Gong <b38343@freescale.com> --- drivers/dma/imx-sdma.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)