| Message ID | 20181019074908.13226-1-boris.brezillon@bootlin.com |
|---|---|
| Headers | show |
| Series | mtd: maps: physmap cleanups | expand |
On Fri, Oct 19, 2018 at 9:49 AM Boris Brezillon <boris.brezillon@bootlin.com> wrote: > This is an attempt at merging physmap_of.c, gpio-addr-flash.c and > physmap.c. The rational behind this decision is code sharing and > consistency with what's done in other parts of the kernel: > > - physmap_of is just adding OF/DT support to the physmap logic, and we > usually put that code in the pre-existing driver instead of creating > a new one > > - gpio-addr-flash is just an extension of the physmap logic which is > needed when the platform does not have enough native ADDR lines > to address the whole flash. Again, I think this core belongs in the > physmap driver > > Patches 1 to 9 are just cleanups, and are not really related to the > merging of physmap_of and gpio-addr-flash into physmap.c. > > Patch 10 to 13 are doing the actual merging work, and patch 14 is > documenting the addr-gpios property that has been added to support > the gpio-addr-flash on DT platforms. > > Patch 15 is a cosmetic change proposed by Ricardo. This series: Acked-by: Linus Walleij <linus.walleij@linaro.org> I expect to test the result when rebasing my patches for Gemini, so hope for this to hit your devel branch after v4.20-rc1. Yours, Linus Walleij
Hi Linus, On Mon, 22 Oct 2018 13:45:49 +0200 Linus Walleij <linus.walleij@linaro.org> wrote: > On Fri, Oct 19, 2018 at 9:49 AM Boris Brezillon > <boris.brezillon@bootlin.com> wrote: > > > This is an attempt at merging physmap_of.c, gpio-addr-flash.c and > > physmap.c. The rational behind this decision is code sharing and > > consistency with what's done in other parts of the kernel: > > > > - physmap_of is just adding OF/DT support to the physmap logic, and we > > usually put that code in the pre-existing driver instead of creating > > a new one > > > > - gpio-addr-flash is just an extension of the physmap logic which is > > needed when the platform does not have enough native ADDR lines > > to address the whole flash. Again, I think this core belongs in the > > physmap driver > > > > Patches 1 to 9 are just cleanups, and are not really related to the > > merging of physmap_of and gpio-addr-flash into physmap.c. > > > > Patch 10 to 13 are doing the actual merging work, and patch 14 is > > documenting the addr-gpios property that has been added to support > > the gpio-addr-flash on DT platforms. > > > > Patch 15 is a cosmetic change proposed by Ricardo. > > This series: > Acked-by: Linus Walleij <linus.walleij@linaro.org> > > I expect to test the result when rebasing my patches for Gemini, > so hope for this to hit your devel branch after v4.20-rc1. > I already have a topic branch [1] in case you want to start working on it before 4.20-rc1. Regards, Boris [1]https://github.com/bbrezillon/linux-0day/tree/mtd/physmap-cleanup
On Fri, 19 Oct 2018 09:48:53 +0200 Boris Brezillon <boris.brezillon@bootlin.com> wrote: > Hello, > > This is an attempt at merging physmap_of.c, gpio-addr-flash.c and > physmap.c. The rational behind this decision is code sharing and > consistency with what's done in other parts of the kernel: > > - physmap_of is just adding OF/DT support to the physmap logic, and we > usually put that code in the pre-existing driver instead of creating > a new one > > - gpio-addr-flash is just an extension of the physmap logic which is > needed when the platform does not have enough native ADDR lines > to address the whole flash. Again, I think this core belongs in the > physmap driver > > Patches 1 to 9 are just cleanups, and are not really related to the > merging of physmap_of and gpio-addr-flash into physmap.c. > > Patch 10 to 13 are doing the actual merging work, and patch 14 is > documenting the addr-gpios property that has been added to support > the gpio-addr-flash on DT platforms. > > Patch 15 is a cosmetic change proposed by Ricardo. > > Regards, > > Boris > > Boris Brezillon (13): > mtd: maps: physmap: Add SPDX header > mtd: maps: physmap: Rename ->map and ->mtd into ->maps and ->mtds > mtd: maps: physmap: Use platform_get_resource() to retrieve iomem > resources > mtd: maps: physmap: Use dev_notice() and a %pR specifier > mtd: maps: physmap: Use devm_ioremap_resource() > mtd: maps: physmap: Remove the MAX_RESOURCES limitation > mtd: maps: physmap: Check mtd_device_{parse_register,unregister}() ret > code > mtd: maps: physmap: Return -ENOMEM directly when info allocation fails > mtd: maps: physmap: Fix coding style issues reported by checkpatch > mtd: maps: Prepare merging of physmap and physmap_of > mtd: maps: Merge physmap_of.c into physmap-core.c > mtd: maps: Rename physmap_of_{versatile,gemini} into > physmap-{versatile,gemini} > mtd: maps: Merge gpio-addr-flash.c into physmap-core.c > > Ricardo Ribalda Delgado (2): > dt-binding: mtd: physmap: Document the addr-gpios property > mtd: maps: physmap: Invert logic on if/else branch Queued to mtd/next.
Hello, This is an attempt at merging physmap_of.c, gpio-addr-flash.c and physmap.c. The rational behind this decision is code sharing and consistency with what's done in other parts of the kernel: - physmap_of is just adding OF/DT support to the physmap logic, and we usually put that code in the pre-existing driver instead of creating a new one - gpio-addr-flash is just an extension of the physmap logic which is needed when the platform does not have enough native ADDR lines to address the whole flash. Again, I think this core belongs in the physmap driver Patches 1 to 9 are just cleanups, and are not really related to the merging of physmap_of and gpio-addr-flash into physmap.c. Patch 10 to 13 are doing the actual merging work, and patch 14 is documenting the addr-gpios property that has been added to support the gpio-addr-flash on DT platforms. Patch 15 is a cosmetic change proposed by Ricardo. Regards, Boris Boris Brezillon (13): mtd: maps: physmap: Add SPDX header mtd: maps: physmap: Rename ->map and ->mtd into ->maps and ->mtds mtd: maps: physmap: Use platform_get_resource() to retrieve iomem resources mtd: maps: physmap: Use dev_notice() and a %pR specifier mtd: maps: physmap: Use devm_ioremap_resource() mtd: maps: physmap: Remove the MAX_RESOURCES limitation mtd: maps: physmap: Check mtd_device_{parse_register,unregister}() ret code mtd: maps: physmap: Return -ENOMEM directly when info allocation fails mtd: maps: physmap: Fix coding style issues reported by checkpatch mtd: maps: Prepare merging of physmap and physmap_of mtd: maps: Merge physmap_of.c into physmap-core.c mtd: maps: Rename physmap_of_{versatile,gemini} into physmap-{versatile,gemini} mtd: maps: Merge gpio-addr-flash.c into physmap-core.c Ricardo Ribalda Delgado (2): dt-binding: mtd: physmap: Document the addr-gpios property mtd: maps: physmap: Invert logic on if/else branch .../devicetree/bindings/mtd/mtd-physmap.txt | 2 + drivers/mtd/maps/Kconfig | 27 +- drivers/mtd/maps/Makefile | 10 +- drivers/mtd/maps/gpio-addr-flash.c | 281 --------- drivers/mtd/maps/physmap-core.c | 655 +++++++++++++++++++++ .../maps/{physmap_of_gemini.c => physmap-gemini.c} | 2 +- .../maps/{physmap_of_gemini.h => physmap-gemini.h} | 2 +- ...{physmap_of_versatile.c => physmap-versatile.c} | 2 +- ...{physmap_of_versatile.h => physmap-versatile.h} | 2 +- drivers/mtd/maps/physmap.c | 280 --------- drivers/mtd/maps/physmap_of_core.c | 368 ------------ 11 files changed, 678 insertions(+), 953 deletions(-) delete mode 100644 drivers/mtd/maps/gpio-addr-flash.c create mode 100644 drivers/mtd/maps/physmap-core.c rename drivers/mtd/maps/{physmap_of_gemini.c => physmap-gemini.c} (98%) rename drivers/mtd/maps/{physmap_of_gemini.h => physmap-gemini.h} (90%) rename drivers/mtd/maps/{physmap_of_versatile.c => physmap-versatile.c} (99%) rename drivers/mtd/maps/{physmap_of_versatile.h => physmap-versatile.h} (90%) delete mode 100644 drivers/mtd/maps/physmap.c delete mode 100644 drivers/mtd/maps/physmap_of_core.c