Linaro-mm-sig

linaro-mm-sig@lists.linaro.org

8 participants
3221 discussions

Re: [Linaro-mm-sig] [PATCH v2] dma_heap: use for_each_sgtable_sg in sg_table release flow

by Robin Murphy

On 2021-11-25 13:49, guangming.cao(a)mediatek.com wrote: > From: Guangming <Guangming.Cao(a)mediatek.com> > > Use (for_each_sgtable_sg) rather than (for_each_sg) to traverse > sg_table to free sg_table. > Use (for_each_sg) maybe will casuse some pages can't be freed > when send wrong nents number. It's still worth spelling out that this is fixing a bug where the current code should have been using table->orig_nents - it's just that switching to the sgtable helper is the best way to make the fix, since it almost entirely removes the possibility of making that (previously rather common) mistake. If it helps, for the change itself: Reviewed-by: Robin Murphy <robin.murphy(a)arm.com> Thanks, Robin. > Signed-off-by: Guangming <Guangming.Cao(a)mediatek.com> > --- > drivers/dma-buf/heaps/system_heap.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/dma-buf/heaps/system_heap.c b/drivers/dma-buf/heaps/system_heap.c > index 23a7e74ef966..8660508f3684 100644 > --- a/drivers/dma-buf/heaps/system_heap.c > +++ b/drivers/dma-buf/heaps/system_heap.c > @@ -289,7 +289,7 @@ static void system_heap_dma_buf_release(struct dma_buf *dmabuf) > int i; > > table = &buffer->sg_table; > - for_each_sg(table->sgl, sg, table->nents, i) { > + for_each_sgtable_sg(table, sg, i) { > struct page *page = sg_page(sg); > > __free_pages(page, compound_order(page)); >

4 years

Re: [Linaro-mm-sig] [PATCH] dma_heap: use sg_table.orig_nents in sg_table release flow

by Robin Murphy

On 2021-11-25 12:46, guangming.cao(a)mediatek.com wrote: > From: Guangming <Guangming.Cao(a)mediatek.com> > > Use (sg_table.orig_nents) rather than (sg_table.nents) to traverse > sg_table to free sg_table. > Use (sg_table.nents) maybe will casuse some pages can't be freed. ...and this sort of bug is precisely why we have the for_each_sgtable_sg() helper ;) Robin. > Signed-off-by: Guangming <Guangming.Cao(a)mediatek.com> > --- > drivers/dma-buf/heaps/system_heap.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/dma-buf/heaps/system_heap.c b/drivers/dma-buf/heaps/system_heap.c > index 23a7e74ef966..ce10d4eb674c 100644 > --- a/drivers/dma-buf/heaps/system_heap.c > +++ b/drivers/dma-buf/heaps/system_heap.c > @@ -289,7 +289,7 @@ static void system_heap_dma_buf_release(struct dma_buf *dmabuf) > int i; > > table = &buffer->sg_table; > - for_each_sg(table->sgl, sg, table->nents, i) { > + for_each_sg(table->sgl, sg, table->orig_nents, i) { > struct page *page = sg_page(sg); > > __free_pages(page, compound_order(page)); >

4 years

Re: [Linaro-mm-sig] [PATCH v3] i2c: tegra: Add the ACPI support

by Dmitry Osipenko

25.11.2021 12:07, Akhil R пишет: > Add support for the ACPI based device registration so that the driver > can be also enabled through ACPI table. > > This does not include the ACPI support for Tegra VI and DVC I2C. > > Signed-off-by: Akhil R <akhilrajeev(a)nvidia.com> > --- > drivers/i2c/busses/i2c-tegra.c | 52 ++++++++++++++++++++++++++++++++---------- > 1 file changed, 40 insertions(+), 12 deletions(-) > > v3 changes > * removed acpi_has_method check. > * moved dev_err_probe to init_reset function to make it consistent with > init_clocks. > * Updates in commit message as suggested. > > v2 - https://lkml.org/lkml/2021/11/23/82 > v1 - https://lkml.org/lkml/2021/11/19/393 Akhil, the patch looks almost good, thank you. Please see one minor question below. > diff --git a/drivers/i2c/busses/i2c-tegra.c b/drivers/i2c/busses/i2c-tegra.c > index c883044..b889eb3 100644 > --- a/drivers/i2c/busses/i2c-tegra.c > +++ b/drivers/i2c/busses/i2c-tegra.c > @@ -6,6 +6,7 @@ > * Author: Colin Cross <ccross(a)android.com> > */ > > +#include <linux/acpi.h> > #include <linux/bitfield.h> > #include <linux/clk.h> > #include <linux/delay.h> > @@ -608,6 +609,7 @@ static int tegra_i2c_wait_for_config_load(struct tegra_i2c_dev *i2c_dev) > static int tegra_i2c_init(struct tegra_i2c_dev *i2c_dev) > { > u32 val, clk_divisor, clk_multiplier, tsu_thd, tlow, thigh, non_hs_mode; > + acpi_handle handle = ACPI_HANDLE(i2c_dev->dev); ... > +static int tegra_i2c_init_reset(struct tegra_i2c_dev *i2c_dev) > +{ > + if (has_acpi_companion(i2c_dev->dev)) > + return 0; Can we use ACPI_HANDLE() everywhere instead of has_acpi_companion()? For consistency. I guess that's what Andy was asking about in v1?

4 years

Re: [Linaro-mm-sig] [PATCH v2] i2c: tegra: Add ACPI support

by Dmitry Osipenko

24.11.2021 19:40, Akhil R пишет: >> 24.11.2021 10:18, Akhil R пишет: >>>> *i2c_dev) >>>>> i2c_dev->is_vi = true; } >>>> How are you going to differentiate the VI I2C from a non-VI? This >>>> doesn't look right. >>> This patch adds the ACPI support to only non-VI I2C. The device_ids in >>> match table are added accordingly. I suppose, of_device_is_compatible >>> always returns false as there is no device tree. >>> Agree with the other comments. >> >> Will the VI I2C have a different ACPI ID or how it's going to work? > As there is a different compatible for VI I2C in device tree, I suppose the ACPI > would have a different ID as well. I think the logic would also need an update > if to have VI I2C using the ACPI. But that wasn't actually considered in this patch. Thanks, you could reflected it in the commit message.

4 years

Re: [Linaro-mm-sig] [PATCH v2] i2c: tegra: Add ACPI support

by Dmitry Osipenko

24.11.2021 10:18, Akhil R пишет: >> *i2c_dev) >>> i2c_dev->is_vi = true; >>> } >> How are you going to differentiate the VI I2C from a non-VI? This doesn't look >> right. > This patch adds the ACPI support to only non-VI I2C. The device_ids in match table > are added accordingly. I suppose, of_device_is_compatible always returns false as > there is no device tree. > Agree with the other comments. Will the VI I2C have a different ACPI ID or how it's going to work?

4 years

Re: [Linaro-mm-sig] [PATCH v2] i2c: tegra: Add ACPI support

by kernel test robot

Hi Akhil, Thank you for the patch! Yet something to improve: [auto build test ERROR on tegra/for-next] [also build test ERROR on v5.16-rc2 next-20211123] [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/Akhil-R/i2c-tegra-Add-ACPI-support… base: https://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux.git for-next config: riscv-buildonly-randconfig-r005-20211123 (https://download.01.org/0day-ci/archive/20211124/202111240017.bYyZ7knz-lkp@…) compiler: clang version 14.0.0 (https://github.com/llvm/llvm-project 49e3838145dff1ec91c2e67a2cb562775c8d2a08) 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 # install riscv cross compiling tool for clang build # apt-get install binutils-riscv64-linux-gnu # https://github.com/0day-ci/linux/commit/dec174be801f41a9e42f4381c59c2357c25… git remote add linux-review https://github.com/0day-ci/linux git fetch --no-tags linux-review Akhil-R/i2c-tegra-Add-ACPI-support/20211123-151636 git checkout dec174be801f41a9e42f4381c59c2357c25e40fb # save the config file to linux build tree COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 ARCH=riscv If you fix the issue, kindly add following tag as appropriate Reported-by: kernel test robot <lkp(a)intel.com> All errors (new ones prefixed by >>): In file included from drivers/i2c/busses/i2c-tegra.c:13: In file included from include/linux/dmaengine.h:12: In file included from include/linux/scatterlist.h:9: In file included from arch/riscv/include/asm/io.h:136: include/asm-generic/io.h:464:31: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic] val = __raw_readb(PCI_IOBASE + addr); ~~~~~~~~~~ ^ include/asm-generic/io.h:477:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic] val = __le16_to_cpu((__le16 __force)__raw_readw(PCI_IOBASE + addr)); ~~~~~~~~~~ ^ include/uapi/linux/byteorder/little_endian.h:36:51: note: expanded from macro '__le16_to_cpu' #define __le16_to_cpu(x) ((__force __u16)(__le16)(x)) ^ In file included from drivers/i2c/busses/i2c-tegra.c:13: In file included from include/linux/dmaengine.h:12: In file included from include/linux/scatterlist.h:9: In file included from arch/riscv/include/asm/io.h:136: include/asm-generic/io.h:490:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic] val = __le32_to_cpu((__le32 __force)__raw_readl(PCI_IOBASE + addr)); ~~~~~~~~~~ ^ include/uapi/linux/byteorder/little_endian.h:34:51: note: expanded from macro '__le32_to_cpu' #define __le32_to_cpu(x) ((__force __u32)(__le32)(x)) ^ In file included from drivers/i2c/busses/i2c-tegra.c:13: In file included from include/linux/dmaengine.h:12: In file included from include/linux/scatterlist.h:9: In file included from arch/riscv/include/asm/io.h:136: include/asm-generic/io.h:501:33: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic] __raw_writeb(value, PCI_IOBASE + addr); ~~~~~~~~~~ ^ include/asm-generic/io.h:511:59: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic] __raw_writew((u16 __force)cpu_to_le16(value), PCI_IOBASE + addr); ~~~~~~~~~~ ^ include/asm-generic/io.h:521:59: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic] __raw_writel((u32 __force)cpu_to_le32(value), PCI_IOBASE + addr); ~~~~~~~~~~ ^ include/asm-generic/io.h:1024:55: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic] return (port > MMIO_UPPER_LIMIT) ? NULL : PCI_IOBASE + port; ~~~~~~~~~~ ^ >> drivers/i2c/busses/i2c-tegra.c:623:16: error: implicit declaration of function 'acpi_has_method' [-Werror,-Wimplicit-function-declaration] if (handle && acpi_has_method(handle, "_RST")) ^ drivers/i2c/busses/i2c-tegra.c:623:16: note: did you mean 'acpi_has_watchdog'? include/linux/acpi.h:1321:20: note: 'acpi_has_watchdog' declared here static inline bool acpi_has_watchdog(void) { return false; } ^ 7 warnings and 1 error generated. vim +/acpi_has_method +623 drivers/i2c/busses/i2c-tegra.c 608 609 static int tegra_i2c_init(struct tegra_i2c_dev *i2c_dev) 610 { 611 u32 val, clk_divisor, clk_multiplier, tsu_thd, tlow, thigh, non_hs_mode; 612 acpi_handle handle = ACPI_HANDLE(i2c_dev->dev); 613 int err; 614 615 /* 616 * The reset shouldn't ever fail in practice. The failure will be a 617 * sign of a severe problem that needs to be resolved. Still we don't 618 * want to fail the initialization completely because this may break 619 * kernel boot up since voltage regulators use I2C. Hence, we will 620 * emit a noisy warning on error, which won't stay unnoticed and 621 * won't hose machine entirely. 622 */ > 623 if (handle && acpi_has_method(handle, "_RST")) 624 err = (acpi_evaluate_object(handle, "_RST", NULL, NULL)); 625 else 626 err = reset_control_reset(i2c_dev->rst); 627 628 WARN_ON_ONCE(err); 629 630 if (i2c_dev->is_dvc) 631 tegra_dvc_init(i2c_dev); 632 633 val = I2C_CNFG_NEW_MASTER_FSM | I2C_CNFG_PACKET_MODE_EN | 634 FIELD_PREP(I2C_CNFG_DEBOUNCE_CNT, 2); 635 636 if (i2c_dev->hw->has_multi_master_mode) 637 val |= I2C_CNFG_MULTI_MASTER_MODE; 638 639 i2c_writel(i2c_dev, val, I2C_CNFG); 640 i2c_writel(i2c_dev, 0, I2C_INT_MASK); 641 642 if (i2c_dev->is_vi) 643 tegra_i2c_vi_init(i2c_dev); 644 645 switch (i2c_dev->bus_clk_rate) { 646 case I2C_MAX_STANDARD_MODE_FREQ + 1 ... I2C_MAX_FAST_MODE_PLUS_FREQ: 647 default: 648 tlow = i2c_dev->hw->tlow_fast_fastplus_mode; 649 thigh = i2c_dev->hw->thigh_fast_fastplus_mode; 650 tsu_thd = i2c_dev->hw->setup_hold_time_fast_fast_plus_mode; 651 652 if (i2c_dev->bus_clk_rate > I2C_MAX_FAST_MODE_FREQ) 653 non_hs_mode = i2c_dev->hw->clk_divisor_fast_plus_mode; 654 else 655 non_hs_mode = i2c_dev->hw->clk_divisor_fast_mode; 656 break; 657 658 case 0 ... I2C_MAX_STANDARD_MODE_FREQ: 659 tlow = i2c_dev->hw->tlow_std_mode; 660 thigh = i2c_dev->hw->thigh_std_mode; 661 tsu_thd = i2c_dev->hw->setup_hold_time_std_mode; 662 non_hs_mode = i2c_dev->hw->clk_divisor_std_mode; 663 break; 664 } 665 666 /* make sure clock divisor programmed correctly */ 667 clk_divisor = FIELD_PREP(I2C_CLK_DIVISOR_HSMODE, 668 i2c_dev->hw->clk_divisor_hs_mode) | 669 FIELD_PREP(I2C_CLK_DIVISOR_STD_FAST_MODE, non_hs_mode); 670 i2c_writel(i2c_dev, clk_divisor, I2C_CLK_DIVISOR); 671 672 if (i2c_dev->hw->has_interface_timing_reg) { 673 val = FIELD_PREP(I2C_INTERFACE_TIMING_THIGH, thigh) | 674 FIELD_PREP(I2C_INTERFACE_TIMING_TLOW, tlow); 675 i2c_writel(i2c_dev, val, I2C_INTERFACE_TIMING_0); 676 } 677 678 /* 679 * Configure setup and hold times only when tsu_thd is non-zero. 680 * Otherwise, preserve the chip default values. 681 */ 682 if (i2c_dev->hw->has_interface_timing_reg && tsu_thd) 683 i2c_writel(i2c_dev, tsu_thd, I2C_INTERFACE_TIMING_1); 684 685 clk_multiplier = (tlow + thigh + 2) * (non_hs_mode + 1); 686 687 err = clk_set_rate(i2c_dev->div_clk, 688 i2c_dev->bus_clk_rate * clk_multiplier); 689 if (err) { 690 dev_err(i2c_dev->dev, "failed to set div-clk rate: %d\n", err); 691 return err; 692 } 693 694 if (!i2c_dev->is_dvc && !i2c_dev->is_vi) { 695 u32 sl_cfg = i2c_readl(i2c_dev, I2C_SL_CNFG); 696 697 sl_cfg |= I2C_SL_CNFG_NACK | I2C_SL_CNFG_NEWSL; 698 i2c_writel(i2c_dev, sl_cfg, I2C_SL_CNFG); 699 i2c_writel(i2c_dev, 0xfc, I2C_SL_ADDR1); 700 i2c_writel(i2c_dev, 0x00, I2C_SL_ADDR2); 701 } 702 703 err = tegra_i2c_flush_fifos(i2c_dev); 704 if (err) 705 return err; 706 707 if (i2c_dev->multimaster_mode && i2c_dev->hw->has_slcg_override_reg) 708 i2c_writel(i2c_dev, I2C_MST_CORE_CLKEN_OVR, I2C_CLKEN_OVERRIDE); 709 710 err = tegra_i2c_wait_for_config_load(i2c_dev); 711 if (err) 712 return err; 713 714 return 0; 715 } 716 --- 0-DAY CI Kernel Test Service, Intel Corporation https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

4 years

Re: [Linaro-mm-sig] [PATCH v2] i2c: tegra: Add ACPI support

by Dmitry Osipenko

23.11.2021 10:15, Akhil R пишет: > Add support for ACPI based device registration so that the driver > can be also enabled through ACPI table. > > Signed-off-by: Akhil R <akhilrajeev(a)nvidia.com> > --- > drivers/i2c/busses/i2c-tegra.c | 52 ++++++++++++++++++++++++++++++++---------- > 1 file changed, 40 insertions(+), 12 deletions(-) > > diff --git a/drivers/i2c/busses/i2c-tegra.c b/drivers/i2c/busses/i2c-tegra.c > index c883044..8e47889 100644 > --- a/drivers/i2c/busses/i2c-tegra.c > +++ b/drivers/i2c/busses/i2c-tegra.c > @@ -6,6 +6,7 @@ > * Author: Colin Cross <ccross(a)android.com> > */ > > +#include <linux/acpi.h> > #include <linux/bitfield.h> > #include <linux/clk.h> > #include <linux/delay.h> > @@ -608,6 +609,7 @@ static int tegra_i2c_wait_for_config_load(struct tegra_i2c_dev *i2c_dev) > static int tegra_i2c_init(struct tegra_i2c_dev *i2c_dev) > { > u32 val, clk_divisor, clk_multiplier, tsu_thd, tlow, thigh, non_hs_mode; > + acpi_handle handle = ACPI_HANDLE(i2c_dev->dev); > int err; > > /* > @@ -618,7 +620,11 @@ static int tegra_i2c_init(struct tegra_i2c_dev *i2c_dev) > * emit a noisy warning on error, which won't stay unnoticed and > * won't hose machine entirely. > */ > - err = reset_control_reset(i2c_dev->rst); > + if (handle && acpi_has_method(handle, "_RST")) Which SoC version doesn't have "_RST" method? If neither, then please remove this check. > + err = (acpi_evaluate_object(handle, "_RST", NULL, NULL)); Please remove parens around acpi_evaluate_object(). Why you added them? > + else > + err = reset_control_reset(i2c_dev->rst); > + > WARN_ON_ONCE(err); > > if (i2c_dev->is_dvc) > @@ -1627,12 +1633,12 @@ static void tegra_i2c_parse_dt(struct tegra_i2c_dev *i2c_dev) > bool multi_mode; > int err; > > - err = of_property_read_u32(np, "clock-frequency", > - &i2c_dev->bus_clk_rate); > + err = device_property_read_u32(i2c_dev->dev, "clock-frequency", > + &i2c_dev->bus_clk_rate); > if (err) > i2c_dev->bus_clk_rate = I2C_MAX_STANDARD_MODE_FREQ; > > - multi_mode = of_property_read_bool(np, "multi-master"); > + multi_mode = device_property_read_bool(i2c_dev->dev, "multi-master"); > i2c_dev->multimaster_mode = multi_mode; > > if (of_device_is_compatible(np, "nvidia,tegra20-i2c-dvc")) > @@ -1642,10 +1648,25 @@ static void tegra_i2c_parse_dt(struct tegra_i2c_dev *i2c_dev) > i2c_dev->is_vi = true; > } How are you going to differentiate the VI I2C from a non-VI? This doesn't look right. > > +static int tegra_i2c_init_reset(struct tegra_i2c_dev *i2c_dev) > +{ > + if (has_acpi_companion(i2c_dev->dev)) > + return 0; > + > + i2c_dev->rst = devm_reset_control_get_exclusive(i2c_dev->dev, "i2c"); > + if (IS_ERR(i2c_dev->rst)) > + return PTR_ERR(i2c_dev->rst); > + > + return 0; > +} > + > static int tegra_i2c_init_clocks(struct tegra_i2c_dev *i2c_dev) > { > int err; > > + if (has_acpi_companion(i2c_dev->dev)) > + return 0; > + > i2c_dev->clocks[i2c_dev->nclocks++].id = "div-clk"; > > if (i2c_dev->hw == &tegra20_i2c_hw || i2c_dev->hw == &tegra30_i2c_hw) > @@ -1720,7 +1741,7 @@ static int tegra_i2c_probe(struct platform_device *pdev) > init_completion(&i2c_dev->msg_complete); > init_completion(&i2c_dev->dma_complete); > > - i2c_dev->hw = of_device_get_match_data(&pdev->dev); > + i2c_dev->hw = device_get_match_data(&pdev->dev); > i2c_dev->cont_id = pdev->id; > i2c_dev->dev = &pdev->dev; > > @@ -1746,15 +1767,13 @@ static int tegra_i2c_probe(struct platform_device *pdev) > if (err) > return err; > > - i2c_dev->rst = devm_reset_control_get_exclusive(i2c_dev->dev, "i2c"); > - if (IS_ERR(i2c_dev->rst)) { > - dev_err_probe(i2c_dev->dev, PTR_ERR(i2c_dev->rst), > - "failed to get reset control\n"); > - return PTR_ERR(i2c_dev->rst); > - } > - > tegra_i2c_parse_dt(i2c_dev); > > + err = tegra_i2c_init_reset(i2c_dev); > + if (err) > + return dev_err_probe(i2c_dev->dev, err, > + "failed to get reset control\n"); This is inconsistent with tegra_i2c_init_clocks() which returns err directly and prints error message within the function. Please move the dev_err_probe() into tegra_i2c_init_reset() to make it consistent, like I suggested before. Please don't reply with a new version of the patch to the old thread, always send a new version separately. Otherwise it's more difficult to follow patches. Lastly, each new version of a patch must contain changelog. I don't see changelog here, please add it to v3 after the " ---" separator of the commit message. Thanks! Example: Commit message. Signed-off-by: Akhil R <akhilrajeev(a)nvidia.com> --- drivers/i2c/busses/i2c-tegra.c | 52 ++++++++++++++++++++++++++++++++---------- 1 file changed, 40 insertions(+), 12 deletions(-) Changelog: v3: bla-bla-bla v2: bla-bla-bla diff ...

4 years

Re: [Linaro-mm-sig] [PATCH v2] i2c: tegra: Add ACPI support

by kernel test robot

4 years

Re: [Linaro-mm-sig] [PATCH 01/15] iio: buffer-dma: Get rid of incoming/outgoing queues

by Lars-Peter Clausen

On 11/22/21 4:16 PM, Paul Cercueil wrote: > Hi Lars, > > Le lun., nov. 22 2021 at 16:08:51 +0100, Lars-Peter Clausen > <lars(a)metafoo.de> a écrit : >> On 11/21/21 9:08 PM, Paul Cercueil wrote: >>> >>> >>> Le dim., nov. 21 2021 at 19:49:03 +0100, Lars-Peter Clausen >>> <lars(a)metafoo.de> a écrit : >>>> On 11/21/21 6:52 PM, Paul Cercueil wrote: >>>>> Hi Lars, >>>>> >>>>> Le dim., nov. 21 2021 at 17:23:35 +0100, Lars-Peter Clausen >>>>> <lars(a)metafoo.de> a écrit : >>>>>> On 11/15/21 3:19 PM, Paul Cercueil wrote: >>>>>>> The buffer-dma code was using two queues, incoming and outgoing, to >>>>>>> manage the state of the blocks in use. >>>>>>> >>>>>>> While this totally works, it adds some complexity to the code, >>>>>>> especially since the code only manages 2 blocks. It is much >>>>>>> easier to >>>>>>> just check each block's state manually, and keep a counter for >>>>>>> the next >>>>>>> block to dequeue. >>>>>>> >>>>>>> Since the new DMABUF based API wouldn't use these incoming and >>>>>>> outgoing >>>>>>> queues anyway, getting rid of them now makes the upcoming changes >>>>>>> simpler. >>>>>>> >>>>>>> Signed-off-by: Paul Cercueil <paul(a)crapouillou.net> >>>>>> The outgoing queue is going to be replaced by fences, but I think >>>>>> we need to keep the incoming queue. >>>>> >>>>> Blocks are always accessed in sequential order, so we now have a >>>>> "queue->next_dequeue" that cycles between the buffers >>>>> allocated for fileio. >>>>> >>>>>>> [...] >>>>>>> @@ -442,28 +435,33 @@ EXPORT_SYMBOL_GPL(iio_dma_buffer_disable); >>>>>>> static void iio_dma_buffer_enqueue(struct iio_dma_buffer_queue >>>>>>> *queue, >>>>>>> struct iio_dma_buffer_block *block) >>>>>>> { >>>>>>> - if (block->state == IIO_BLOCK_STATE_DEAD) { >>>>>>> + if (block->state == IIO_BLOCK_STATE_DEAD) >>>>>>> iio_buffer_block_put(block); >>>>>>> - } else if (queue->active) { >>>>>>> + else if (queue->active) >>>>>>> iio_dma_buffer_submit_block(queue, block); >>>>>>> - } else { >>>>>>> + else >>>>>>> block->state = IIO_BLOCK_STATE_QUEUED; >>>>>>> - list_add_tail(&block->head, &queue->incoming); >>>>>> If iio_dma_buffer_enqueue() is called with a dmabuf and the >>>>>> buffer is not active, it will be marked as queued, but we >>>>>> don't actually keep a reference to it anywhere. It will >>>>>> never be submitted to the DMA, and it will never be >>>>>> signaled as completed. >>>>> >>>>> We do keep a reference to the buffers, in the queue->fileio.blocks >>>>> array. When the buffer is enabled, all the blocks in that >>>>> array that are in the "queued" state will be submitted to the >>>>> DMA. >>>>> >>>> But not when used in combination with the DMA buf changes later in >>>> this series. >>>> >>> >>> That's still the case after the DMABUF changes of the series. Or can >>> you point me exactly what you think is broken? >>> >> When you allocate a DMABUF with the allocate IOCTL and then submit it >> with the enqueue IOCTL before the buffer is enabled it will end up >> marked as queued, but not actually be queued anywhere. >> > > Ok, it works for me because I never enqueue blocks before enabling the > buffer. I can add a requirement that blocks must be enqueued only > after the buffer is enabled. I don't think that is a good idea. This way you are going to potentially drop data at the begining of your stream when the DMA isn't ready yet.

4 years

Re: [Linaro-mm-sig] [PATCH 01/15] iio: buffer-dma: Get rid of incoming/outgoing queues

by Lars-Peter Clausen

On 11/21/21 9:08 PM, Paul Cercueil wrote: > > > Le dim., nov. 21 2021 at 19:49:03 +0100, Lars-Peter Clausen > <lars(a)metafoo.de> a écrit : >> On 11/21/21 6:52 PM, Paul Cercueil wrote: >>> Hi Lars, >>> >>> Le dim., nov. 21 2021 at 17:23:35 +0100, Lars-Peter Clausen >>> <lars(a)metafoo.de> a écrit : >>>> On 11/15/21 3:19 PM, Paul Cercueil wrote: >>>>> The buffer-dma code was using two queues, incoming and outgoing, to >>>>> manage the state of the blocks in use. >>>>> >>>>> While this totally works, it adds some complexity to the code, >>>>> especially since the code only manages 2 blocks. It is much easier to >>>>> just check each block's state manually, and keep a counter for the >>>>> next >>>>> block to dequeue. >>>>> >>>>> Since the new DMABUF based API wouldn't use these incoming and >>>>> outgoing >>>>> queues anyway, getting rid of them now makes the upcoming changes >>>>> simpler. >>>>> >>>>> Signed-off-by: Paul Cercueil <paul(a)crapouillou.net> >>>> The outgoing queue is going to be replaced by fences, but I think >>>> we need to keep the incoming queue. >>> >>> Blocks are always accessed in sequential order, so we now have a >>> "queue->next_dequeue" that cycles between the buffers allocated for >>> fileio. >>> >>>>> [...] >>>>> @@ -442,28 +435,33 @@ EXPORT_SYMBOL_GPL(iio_dma_buffer_disable); >>>>> static void iio_dma_buffer_enqueue(struct iio_dma_buffer_queue >>>>> *queue, >>>>> struct iio_dma_buffer_block *block) >>>>> { >>>>> - if (block->state == IIO_BLOCK_STATE_DEAD) { >>>>> + if (block->state == IIO_BLOCK_STATE_DEAD) >>>>> iio_buffer_block_put(block); >>>>> - } else if (queue->active) { >>>>> + else if (queue->active) >>>>> iio_dma_buffer_submit_block(queue, block); >>>>> - } else { >>>>> + else >>>>> block->state = IIO_BLOCK_STATE_QUEUED; >>>>> - list_add_tail(&block->head, &queue->incoming); >>>> If iio_dma_buffer_enqueue() is called with a dmabuf and the buffer >>>> is not active, it will be marked as queued, but we don't actually >>>> keep a reference to it anywhere. It will never be submitted to >>>> the DMA, and it will never be signaled as completed. >>> >>> We do keep a reference to the buffers, in the queue->fileio.blocks >>> array. When the buffer is enabled, all the blocks in that array >>> that are in the "queued" state will be submitted to the DMA. >>> >> But not when used in combination with the DMA buf changes later in >> this series. >> > > That's still the case after the DMABUF changes of the series. Or can > you point me exactly what you think is broken? > When you allocate a DMABUF with the allocate IOCTL and then submit it with the enqueue IOCTL before the buffer is enabled it will end up marked as queued, but not actually be queued anywhere.

4 years

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

Linaro-mm-sig