mbox series

Please pull mmc mmc-2-19-2020

Message ID AM0PR04MB448135DEF747DDE2D4AEBB3E88130@AM0PR04MB4481.eurprd04.prod.outlook.com
State Changes Requested
Delegated to: Tom Rini
Headers show
Series Please pull mmc mmc-2-19-2020 | expand

Pull-request

https://gitlab.denx.de/u-boot/custodians/u-boot-mmc.git tags/mmc-2-19-2020

Message

Peng Fan Feb. 20, 2020, 1:56 a.m. UTC
Hi Tom

Please pull mmc-2-29-2020.

------------------------------------
sdhci: code clean-up and fix cache coherency problem
enable cache snooping on mpc830x
Fix build error when MMC_WRITE disabled
------------------------------------

CI: https://travis-ci.org/MrVan/u-boot/builds/652476827

Thanks,
Peng

The following changes since commit f2a73d6867ef973fbb8471cc87058205999b5e5c:

  Merge tag 'u-boot-stm32-20200214' of https://gitlab.denx.de/u-boot/custodians/u-boot-stm (2020-02-14 07:31:47 -0500)

are available in the Git repository at:

  https://gitlab.denx.de/u-boot/custodians/u-boot-mmc.git tags/mmc-2-19-2020

for you to fetch changes up to 5cc25dcd5d79ef85fdad807a3e5a7af10333cfc3:

  mmc: sdhci: fix missing cache invalidation after reading by DMA (2020-02-19 21:27:31 +0800)

----------------------------------------------------------------
Bharat Kumar Reddy Gooty (1):
      drivers: mmc: rpmb: Use R1 response

Jaehoon Chung (1):
      mmc: fix the build error when MMC_WRITE is disabled

Masahiro Yamada (14):
      mmc: sdhci-cadence: send tune request twice to work around errata
      mmc: check the return value of mmc_select_mode_and_width()
      mmc: remove unneeded forward declarations
      dma-mapping: fix the prototype of dma_map_single()
      dma-mapping: fix the prototype of dma_unmap_single()
      dma-mapping: move dma_map_(un)single() to <linux/dma-mapping.h>
      dma-mapping: add <asm/dma-mapping.h> for all architectures
      mmc: sdhci: put the aligned buffer pointer to struct sdhci_host
      mmc: sdhci: reduce code duplication for aligned buffer
      mmc: sdhci: use lower_32_bit2() and upper_32_bits() for setting adma_addr
      mmc: sdhci: remove unneeded casts
      mmc: add mmc_get_dma_dir() helper
      mmc: sdhci: use dma_map_single() instead of flush_cache() before DMA
      mmc: sdhci: fix missing cache invalidation after reading by DMA

Rasmus Villemoes (1):
      mmc: fsl_esdhc: actually enable cache snooping on mpc830x

 arch/arc/include/asm/dma-mapping.h        |  1 +
 arch/arm/include/asm/dma-mapping.h        | 29 +----------------------------
 arch/m68k/include/asm/dma-mapping.h       |  1 +
 arch/microblaze/include/asm/dma-mapping.h |  1 +
 arch/mips/include/asm/dma-mapping.h       |  1 +
 arch/nds32/include/asm/dma-mapping.h      | 27 +--------------------------
 arch/nios2/include/asm/dma-mapping.h      | 25 +------------------------
 arch/powerpc/include/asm/dma-mapping.h    |  1 +
 arch/riscv/include/asm/dma-mapping.h      | 29 +----------------------------
 arch/sandbox/include/asm/dma-mapping.h    |  1 +
 arch/sh/include/asm/dma-mapping.h         |  1 +
 arch/x86/include/asm/dma-mapping.h        | 29 +----------------------------
 arch/xtensa/include/asm/dma-mapping.h     |  1 +
 drivers/dma/ti/k3-udma.c                  |  2 +-
 drivers/mmc/fsl_esdhc.c                   | 15 +++++++++++++--
 drivers/mmc/mmc.c                         |  8 +++-----
 drivers/mmc/rpmb.c                        |  5 +++++
 drivers/mmc/sdhci-cadence.c               | 21 +++++++++++++++++----
 drivers/mmc/sdhci.c                       | 96 ++++++++++++++++++++++++++++++++++++++++++++++--------------------------------------------------
 drivers/mmc/tmio-common.c                 |  5 ++---
 drivers/mtd/nand/raw/denali.c             |  5 ++---
 drivers/net/altera_tse.c                  |  2 +-
 drivers/net/ftmac110.c                    |  2 +-
 drivers/net/macb.c                        |  4 ++--
 drivers/soc/ti/k3-navss-ringacc.c         |  2 +-
 drivers/ufs/ufs.c                         |  2 +-
 drivers/usb/cdns3/gadget.c                |  2 +-
 drivers/usb/dwc3/core.c                   |  8 ++++----
 drivers/usb/dwc3/gadget.c                 |  2 +-
 drivers/usb/gadget/udc/udc-core.c         |  4 ++--
 include/linux/dma-mapping.h               | 63 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 include/mmc.h                             |  6 ++++++
 include/sdhci.h                           |  3 +++
 33 files changed, 188 insertions(+), 216 deletions(-)
 create mode 100644 arch/arc/include/asm/dma-mapping.h
 create mode 100644 arch/m68k/include/asm/dma-mapping.h
 create mode 100644 arch/microblaze/include/asm/dma-mapping.h
 create mode 100644 arch/mips/include/asm/dma-mapping.h
 create mode 100644 arch/powerpc/include/asm/dma-mapping.h
 create mode 100644 arch/sandbox/include/asm/dma-mapping.h
 create mode 100644 arch/sh/include/asm/dma-mapping.h
 create mode 100644 arch/xtensa/include/asm/dma-mapping.h
 create mode 100644 include/linux/dma-mapping.h

Comments

Tom Rini Feb. 20, 2020, 4:33 a.m. UTC | #1
On Thu, Feb 20, 2020 at 01:56:41AM +0000, Peng Fan wrote:

> Hi Tom
> 
> Please pull mmc-2-29-2020.
> 

NAK, this breaks nios2:
     nios2:  +   10m50 3c120
+(10m50,3c120) ../drivers/net/altera_tse.c: In function 'altera_tse_probe':
+(10m50,3c120) ../drivers/net/altera_tse.c:646:15: error: implicit declaration of function 'dma_alloc_coherent'; did you mean 'lmb_alloc_addr'? [-Werror=implicit-function-declaration]
+(10m50,3c120)     desc_mem = dma_alloc_coherent(len, &addr);
+(10m50,3c120)                ^~~~~~~~~~~~~~~~~~
+(10m50,3c120)                lmb_alloc_addr
+(10m50,3c120) ../drivers/net/altera_tse.c:646:13: error: assignment makes pointer from integer without a cast [-Werror=int-conversion]
+(10m50,3c120)              ^
+(10m50,3c120) cc1: all warnings being treated as errors
+(10m50,3c120) make[2]: *** [drivers/net/altera_tse.o] Error 1
+(10m50,3c120) make[1]: *** [drivers/net] Error 2
+(10m50,3c120) make: *** [sub-make] Error 2
Masahiro Yamada Feb. 20, 2020, 5:18 a.m. UTC | #2
Hi Peng,

On Thu, Feb 20, 2020 at 1:34 PM Tom Rini <trini@konsulko.com> wrote:
>
> On Thu, Feb 20, 2020 at 01:56:41AM +0000, Peng Fan wrote:
>
> > Hi Tom
> >
> > Please pull mmc-2-29-2020.
> >
>
> NAK, this breaks nios2:
>      nios2:  +   10m50 3c120
> +(10m50,3c120) ../drivers/net/altera_tse.c: In function 'altera_tse_probe':
> +(10m50,3c120) ../drivers/net/altera_tse.c:646:15: error: implicit declaration of function 'dma_alloc_coherent'; did you mean 'lmb_alloc_addr'? [-Werror=implicit-function-declaration]
> +(10m50,3c120)     desc_mem = dma_alloc_coherent(len, &addr);
> +(10m50,3c120)                ^~~~~~~~~~~~~~~~~~
> +(10m50,3c120)                lmb_alloc_addr
> +(10m50,3c120) ../drivers/net/altera_tse.c:646:13: error: assignment makes pointer from integer without a cast [-Werror=int-conversion]
> +(10m50,3c120)              ^
> +(10m50,3c120) cc1: all warnings being treated as errors
> +(10m50,3c120) make[2]: *** [drivers/net/altera_tse.o] Error 1
> +(10m50,3c120) make[1]: *** [drivers/net] Error 2
> +(10m50,3c120) make: *** [sub-make] Error 2
>
> --
> Tom


Agh, something strange happens somehow in the following commit.

commit 87a6590d6eb500db95bcdb09c16efb16559f67c6
Author: Masahiro Yamada <yamada.masahiro@socionext.com>
Date:   Fri Feb 14 16:40:20 2020 +0900

    dma-mapping: add <asm/dma-mapping.h> for all architectures



Peng,

Could you replace it with this?
http://patchwork.ozlabs.org/patch/1241209/

(Or, just revert the diff for arch/nios2/include/asm/dma-mapping)
Peng Fan Feb. 20, 2020, 6:52 a.m. UTC | #3
> Subject: Re: Please pull mmc mmc-2-19-2020
> 
> Hi Peng,
> 
> On Thu, Feb 20, 2020 at 1:34 PM Tom Rini <trini@konsulko.com> wrote:
> >
> > On Thu, Feb 20, 2020 at 01:56:41AM +0000, Peng Fan wrote:
> >
> > > Hi Tom
> > >
> > > Please pull mmc-2-29-2020.
> > >
> >
> > NAK, this breaks nios2:
> >      nios2:  +   10m50 3c120
> > +(10m50,3c120) ../drivers/net/altera_tse.c: In function 'altera_tse_probe':
> > +(10m50,3c120) ../drivers/net/altera_tse.c:646:15: error: implicit
> declaration of function 'dma_alloc_coherent'; did you mean 'lmb_alloc_addr'?
> [-Werror=implicit-function-declaration]
> > +(10m50,3c120)     desc_mem = dma_alloc_coherent(len, &addr);
> > +(10m50,3c120)                ^~~~~~~~~~~~~~~~~~
> > +(10m50,3c120)                lmb_alloc_addr
> > +(10m50,3c120) ../drivers/net/altera_tse.c:646:13: error: assignment
> makes pointer from integer without a cast [-Werror=int-conversion]
> > +(10m50,3c120)              ^
> > +(10m50,3c120) cc1: all warnings being treated as errors
> > +(10m50,3c120) make[2]: *** [drivers/net/altera_tse.o] Error 1
> > +(10m50,3c120) make[1]: *** [drivers/net] Error 2
> > +(10m50,3c120) make: *** [sub-make] Error 2

I not met such issue in github ci.

> >
> > --
> > Tom
> 
> 
> Agh, something strange happens somehow in the following commit.
> 
> commit 87a6590d6eb500db95bcdb09c16efb16559f67c6
> Author: Masahiro Yamada <yamada.masahiro@socionext.com>
> Date:   Fri Feb 14 16:40:20 2020 +0900
> 
>     dma-mapping: add <asm/dma-mapping.h> for all architectures
> 
> 
> 
> Peng,
> 
> Could you replace it with this?

Replaced. If there still build issue, I'll drop this single patch in this pull request.

Thanks,
Peng.

> https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fpatchw
> ork.ozlabs.org%2Fpatch%2F1241209%2F&amp;data=02%7C01%7Cpeng.fan%
> 40nxp.com%7C7dfca232910b461e892508d7b5c46682%7C686ea1d3bc2b4c6
> fa92cd99c5c301635%7C0%7C0%7C637177727450297022&amp;sdata=bwE9
> pa9V80pmSD6%2FP4vMfBNt8xkJ%2B8R60%2FwM1xkhNGY%3D&amp;reserv
> ed=0
> 
> (Or, just revert the diff for arch/nios2/include/asm/dma-mapping)
> 
> 
> 
> 
> --
> Best Regards
> Masahiro Yamada
Masahiro Yamada Feb. 20, 2020, 7:22 a.m. UTC | #4
Hi Peng,



On Thu, Feb 20, 2020 at 3:52 PM Peng Fan <peng.fan@nxp.com> wrote:
>
> > Subject: Re: Please pull mmc mmc-2-19-2020
> >
> > Hi Peng,
> >
> > On Thu, Feb 20, 2020 at 1:34 PM Tom Rini <trini@konsulko.com> wrote:
> > >
> > > On Thu, Feb 20, 2020 at 01:56:41AM +0000, Peng Fan wrote:
> > >
> > > > Hi Tom
> > > >
> > > > Please pull mmc-2-29-2020.
> > > >
> > >
> > > NAK, this breaks nios2:
> > >      nios2:  +   10m50 3c120
> > > +(10m50,3c120) ../drivers/net/altera_tse.c: In function 'altera_tse_probe':
> > > +(10m50,3c120) ../drivers/net/altera_tse.c:646:15: error: implicit
> > declaration of function 'dma_alloc_coherent'; did you mean 'lmb_alloc_addr'?
> > [-Werror=implicit-function-declaration]
> > > +(10m50,3c120)     desc_mem = dma_alloc_coherent(len, &addr);
> > > +(10m50,3c120)                ^~~~~~~~~~~~~~~~~~
> > > +(10m50,3c120)                lmb_alloc_addr
> > > +(10m50,3c120) ../drivers/net/altera_tse.c:646:13: error: assignment
> > makes pointer from integer without a cast [-Werror=int-conversion]
> > > +(10m50,3c120)              ^
> > > +(10m50,3c120) cc1: all warnings being treated as errors
> > > +(10m50,3c120) make[2]: *** [drivers/net/altera_tse.o] Error 1
> > > +(10m50,3c120) make[1]: *** [drivers/net] Error 2
> > > +(10m50,3c120) make: *** [sub-make] Error 2
>
> I not met such issue in github ci.
>
> > >
> > > --
> > > Tom
> >
> >
> > Agh, something strange happens somehow in the following commit.
> >
> > commit 87a6590d6eb500db95bcdb09c16efb16559f67c6
> > Author: Masahiro Yamada <yamada.masahiro@socionext.com>
> > Date:   Fri Feb 14 16:40:20 2020 +0900
> >
> >     dma-mapping: add <asm/dma-mapping.h> for all architectures
> >
> >
> >
> > Peng,
> >
> > Could you replace it with this?
>
> Replaced. If there still build issue, I'll drop this single patch in this pull request.


No, you cannot do it.

If you dropped it, the following error would come back again.
https://lists.denx.de/pipermail/u-boot/2020-February/400180.html


I confirmed nios2 builds passed, so I hope it will be OK this time.


--
Best Regards
Masahiro Yamada
Tom Rini Feb. 20, 2020, 1:28 p.m. UTC | #5
On Thu, Feb 20, 2020 at 06:52:18AM +0000, Peng Fan wrote:
> > Subject: Re: Please pull mmc mmc-2-19-2020
> > 
> > Hi Peng,
> > 
> > On Thu, Feb 20, 2020 at 1:34 PM Tom Rini <trini@konsulko.com> wrote:
> > >
> > > On Thu, Feb 20, 2020 at 01:56:41AM +0000, Peng Fan wrote:
> > >
> > > > Hi Tom
> > > >
> > > > Please pull mmc-2-29-2020.
> > > >
> > >
> > > NAK, this breaks nios2:
> > >      nios2:  +   10m50 3c120
> > > +(10m50,3c120) ../drivers/net/altera_tse.c: In function 'altera_tse_probe':
> > > +(10m50,3c120) ../drivers/net/altera_tse.c:646:15: error: implicit
> > declaration of function 'dma_alloc_coherent'; did you mean 'lmb_alloc_addr'?
> > [-Werror=implicit-function-declaration]
> > > +(10m50,3c120)     desc_mem = dma_alloc_coherent(len, &addr);
> > > +(10m50,3c120)                ^~~~~~~~~~~~~~~~~~
> > > +(10m50,3c120)                lmb_alloc_addr
> > > +(10m50,3c120) ../drivers/net/altera_tse.c:646:13: error: assignment
> > makes pointer from integer without a cast [-Werror=int-conversion]
> > > +(10m50,3c120)              ^
> > > +(10m50,3c120) cc1: all warnings being treated as errors
> > > +(10m50,3c120) make[2]: *** [drivers/net/altera_tse.o] Error 1
> > > +(10m50,3c120) make[1]: *** [drivers/net] Error 2
> > > +(10m50,3c120) make: *** [sub-make] Error 2
> 
> I not met such issue in github ci.

Travis is indeed missing nios2 tests but it's in GitLab / Azure.