| Message ID | 20210716182241.218705-1-marex@denx.de (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | dmaengine: xilinx: Add empty device_config function | expand |
On 16-07-21, 20:22, Marek Vasut wrote: > Various DMA users call the dmaengine_slave_config() and expect it to > succeed, but that can only succeed if .device_config is implemented. > Add empty device_config function rather than patching all the places > which use dmaengine_slave_config(). .device_config is optional, Yes the dmaengine_slave_config() will check and return error... I think it would make sense to handle this in caller... (ignore ENOSYS..) rather than add a dummy one > > Signed-off-by: Marek Vasut <marex@denx.de> > Cc: Akinobu Mita <akinobu.mita@gmail.com> > Cc: Kedareswara rao Appana <appana.durga.rao@xilinx.com> > Cc: Michal Simek <monstr@monstr.eu> > Cc: Vinod Koul <vinod.koul@intel.com> ummm..? you really need to update this :) > --- > drivers/dma/xilinx/xilinx_dma.c | 12 ++++++++++++ > 1 file changed, 12 insertions(+) > > diff --git a/drivers/dma/xilinx/xilinx_dma.c b/drivers/dma/xilinx/xilinx_dma.c > index 4b9530a7bf65..d6f4bf0d50e8 100644 > --- a/drivers/dma/xilinx/xilinx_dma.c > +++ b/drivers/dma/xilinx/xilinx_dma.c > @@ -1658,6 +1658,17 @@ static void xilinx_dma_issue_pending(struct dma_chan *dchan) > spin_unlock_irqrestore(&chan->lock, flags); > } > > +/** > + * xilinx_dma_issue_pending - Configure the DMA channel > + * @dchan: DMA channel > + * @config: channel configuration > + */ > +static int xilinx_dma_device_config(struct dma_chan *dchan, > + struct dma_slave_config *config) > +{ > + return 0; > +} > + > /** > * xilinx_dma_complete_descriptor - Mark the active descriptor as complete > * @chan : xilinx DMA channel > @@ -3096,6 +3107,7 @@ static int xilinx_dma_probe(struct platform_device *pdev) > xdev->common.device_synchronize = xilinx_dma_synchronize; > xdev->common.device_tx_status = xilinx_dma_tx_status; > xdev->common.device_issue_pending = xilinx_dma_issue_pending; > + xdev->common.device_config = xilinx_dma_device_config; > if (xdev->dma_config->dmatype == XDMA_TYPE_AXIDMA) { > dma_cap_set(DMA_CYCLIC, xdev->common.cap_mask); > xdev->common.device_prep_slave_sg = xilinx_dma_prep_slave_sg; > -- > 2.30.2
On 7/17/21 1:36 PM, Vinod Koul wrote: > On 16-07-21, 20:22, Marek Vasut wrote: >> Various DMA users call the dmaengine_slave_config() and expect it to >> succeed, but that can only succeed if .device_config is implemented. >> Add empty device_config function rather than patching all the places >> which use dmaengine_slave_config(). > > .device_config is optional, Yes the dmaengine_slave_config() will check > and return error... > > I think it would make sense to handle this in caller... (ignore > ENOSYS..) rather than add a dummy one That's what I was trying to avoid -- patching all the places in kernel which might fail. Why handle it in caller ? >> Signed-off-by: Marek Vasut <marex@denx.de> >> Cc: Akinobu Mita <akinobu.mita@gmail.com> >> Cc: Kedareswara rao Appana <appana.durga.rao@xilinx.com> >> Cc: Michal Simek <monstr@monstr.eu> >> Cc: Vinod Koul <vinod.koul@intel.com> > > ummm..? you really need to update this :) I had the patch around for a while indeed, it fell through the cracks.
On 17-07-21, 14:01, Marek Vasut wrote: > On 7/17/21 1:36 PM, Vinod Koul wrote: > > On 16-07-21, 20:22, Marek Vasut wrote: > > > Various DMA users call the dmaengine_slave_config() and expect it to > > > succeed, but that can only succeed if .device_config is implemented. > > > Add empty device_config function rather than patching all the places > > > which use dmaengine_slave_config(). > > > > .device_config is optional, Yes the dmaengine_slave_config() will check > > and return error... > > > > I think it would make sense to handle this in caller... (ignore > > ENOSYS..) rather than add a dummy one > > That's what I was trying to avoid -- patching all the places in kernel which > might fail. Why handle it in caller ? And how many places would that be..? The xilinx driver using xilinx dma right> > > > Signed-off-by: Marek Vasut <marex@denx.de> > > > Cc: Akinobu Mita <akinobu.mita@gmail.com> > > > Cc: Kedareswara rao Appana <appana.durga.rao@xilinx.com> > > > Cc: Michal Simek <monstr@monstr.eu> > > > Cc: Vinod Koul <vinod.koul@intel.com> > > > > ummm..? you really need to update this :) > > I had the patch around for a while indeed, it fell through the cracks. This has to be more than 3 yrs old then!
On 7/17/21 4:34 PM, Vinod Koul wrote: > On 17-07-21, 14:01, Marek Vasut wrote: >> On 7/17/21 1:36 PM, Vinod Koul wrote: >>> On 16-07-21, 20:22, Marek Vasut wrote: >>>> Various DMA users call the dmaengine_slave_config() and expect it to >>>> succeed, but that can only succeed if .device_config is implemented. >>>> Add empty device_config function rather than patching all the places >>>> which use dmaengine_slave_config(). >>> >>> .device_config is optional, Yes the dmaengine_slave_config() will check >>> and return error... >>> >>> I think it would make sense to handle this in caller... (ignore >>> ENOSYS..) rather than add a dummy one >> >> That's what I was trying to avoid -- patching all the places in kernel which >> might fail. Why handle it in caller ? > > And how many places would that be..? The xilinx driver using xilinx > dma right> git grep indicates around 170 matches on dmaengine_slave_config. In my case, it is generic PCM DMA in sound/soc/soc-generic-dmaengine-pcm.c . >>>> Signed-off-by: Marek Vasut <marex@denx.de> >>>> Cc: Akinobu Mita <akinobu.mita@gmail.com> >>>> Cc: Kedareswara rao Appana <appana.durga.rao@xilinx.com> >>>> Cc: Michal Simek <monstr@monstr.eu> >>>> Cc: Vinod Koul <vinod.koul@intel.com> >>> >>> ummm..? you really need to update this :) >> >> I had the patch around for a while indeed, it fell through the cracks. > > This has to be more than 3 yrs old then! Four
Hi Marek,
I love your patch! Perhaps something to improve:
[auto build test WARNING on v5.14-rc1]
[also build test WARNING on next-20210716]
[cannot apply to xlnx/master]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]
url: https://github.com/0day-ci/linux/commits/Marek-Vasut/dmaengine-xilinx-Add-empty-device_config-function/20210718-111119
base: e73f0f0ee7541171d89f2e2491130c7771ba58d3
config: arm-defconfig (attached as .config)
compiler: arm-linux-gnueabi-gcc (GCC) 10.3.0
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# https://github.com/0day-ci/linux/commit/89cc716182246c22a6c7bdb666952b3cec945a0a
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Marek-Vasut/dmaengine-xilinx-Add-empty-device_config-function/20210718-111119
git checkout 89cc716182246c22a6c7bdb666952b3cec945a0a
# save the attached .config to linux build tree
mkdir build_dir
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-10.3.0 make.cross O=build_dir ARCH=arm SHELL=/bin/bash drivers/dma/xilinx/
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
All warnings (new ones prefixed by >>):
drivers/dma/xilinx/xilinx_dma.c:800: warning: expecting prototype for xilinx_dma_tx_descriptor(). Prototype was for xilinx_dma_alloc_tx_descriptor() instead
>> drivers/dma/xilinx/xilinx_dma.c:1659: warning: expecting prototype for xilinx_dma_issue_pending(). Prototype was for xilinx_dma_device_config() instead
drivers/dma/xilinx/xilinx_dma.c:2489: warning: expecting prototype for xilinx_dma_channel_set_config(). Prototype was for xilinx_vdma_channel_set_config() instead
vim +1659 drivers/dma/xilinx/xilinx_dma.c
1651
1652 /**
1653 * xilinx_dma_issue_pending - Configure the DMA channel
1654 * @dchan: DMA channel
1655 * @config: channel configuration
1656 */
1657 static int xilinx_dma_device_config(struct dma_chan *dchan,
1658 struct dma_slave_config *config)
> 1659 {
1660 return 0;
1661 }
1662
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
On 17-07-21, 16:48, Marek Vasut wrote: > On 7/17/21 4:34 PM, Vinod Koul wrote: > > On 17-07-21, 14:01, Marek Vasut wrote: > > > On 7/17/21 1:36 PM, Vinod Koul wrote: > > > > On 16-07-21, 20:22, Marek Vasut wrote: > > > > > Various DMA users call the dmaengine_slave_config() and expect it to > > > > > succeed, but that can only succeed if .device_config is implemented. > > > > > Add empty device_config function rather than patching all the places > > > > > which use dmaengine_slave_config(). > > > > > > > > .device_config is optional, Yes the dmaengine_slave_config() will check > > > > and return error... > > > > > > > > I think it would make sense to handle this in caller... (ignore > > > > ENOSYS..) rather than add a dummy one > > > > > > That's what I was trying to avoid -- patching all the places in kernel which > > > might fail. Why handle it in caller ? > > > > And how many places would that be..? The xilinx driver using xilinx > > dma right> > > git grep indicates around 170 matches on dmaengine_slave_config. In my case, > it is generic PCM DMA in sound/soc/soc-generic-dmaengine-pcm.c . Okay lets have this. Looks like kbuild-bot is not happy, can you fix that and send update
diff --git a/drivers/dma/xilinx/xilinx_dma.c b/drivers/dma/xilinx/xilinx_dma.c index 4b9530a7bf65..d6f4bf0d50e8 100644 --- a/drivers/dma/xilinx/xilinx_dma.c +++ b/drivers/dma/xilinx/xilinx_dma.c @@ -1658,6 +1658,17 @@ static void xilinx_dma_issue_pending(struct dma_chan *dchan) spin_unlock_irqrestore(&chan->lock, flags); } +/** + * xilinx_dma_issue_pending - Configure the DMA channel + * @dchan: DMA channel + * @config: channel configuration + */ +static int xilinx_dma_device_config(struct dma_chan *dchan, + struct dma_slave_config *config) +{ + return 0; +} + /** * xilinx_dma_complete_descriptor - Mark the active descriptor as complete * @chan : xilinx DMA channel @@ -3096,6 +3107,7 @@ static int xilinx_dma_probe(struct platform_device *pdev) xdev->common.device_synchronize = xilinx_dma_synchronize; xdev->common.device_tx_status = xilinx_dma_tx_status; xdev->common.device_issue_pending = xilinx_dma_issue_pending; + xdev->common.device_config = xilinx_dma_device_config; if (xdev->dma_config->dmatype == XDMA_TYPE_AXIDMA) { dma_cap_set(DMA_CYCLIC, xdev->common.cap_mask); xdev->common.device_prep_slave_sg = xilinx_dma_prep_slave_sg;
Various DMA users call the dmaengine_slave_config() and expect it to succeed, but that can only succeed if .device_config is implemented. Add empty device_config function rather than patching all the places which use dmaengine_slave_config(). Signed-off-by: Marek Vasut <marex@denx.de> Cc: Akinobu Mita <akinobu.mita@gmail.com> Cc: Kedareswara rao Appana <appana.durga.rao@xilinx.com> Cc: Michal Simek <monstr@monstr.eu> Cc: Vinod Koul <vinod.koul@intel.com> --- drivers/dma/xilinx/xilinx_dma.c | 12 ++++++++++++ 1 file changed, 12 insertions(+)