Message ID | 20190617073252.27810-1-jagan@amarulasolutions.com |
---|---|
Headers | show |
Series |
|
Related | show |
On Mon, Jun 17, 2019 at 12:37 AM Jagan Teki <jagan@amarulasolutions.com> wrote: > > This is the v2 set for supporting LPDDR4 with associated > features, wrt to previous series[1]. > > Thanks to > - YouMin Chen > - Akash Gajjar > - Kever Yang > for supporting all the help on this work. > > On summary this series support > - Code warning and fixes > - rank detection, this would required to probe single channel > sdram configured in NanoPI-NEO4 > - LPDDR4 support, tested in Rockpro64 and Rock-PI-4 > > Changes for v2: > - handle LPDDR4 code as part of CONFIG_RAM_RK3399_LPDDR4 > - support data_training and set_rate via sdram_rk3399_ops > - add proper sys_reg_enc macros > - add new patch to rename variable sdram_params with params > - fix few commit messages > > patch 0001 - 0034: fix code warnings, prints, new macros > > patch 0035 - 0052: rank detection, sdram debug code > > patch 0053: use DDR3-1800 on NanoPI-NEO4 > > patch 0054 - 0094: lpddr4 support > > patch 0095: enable lpddr4 in Rockpro64 > > patch 0096: enable lpddr4 in Rock-PI-4 > > patch 0097: LPDDR4-100 timings > > patch 0098: Use LPDDR4-100 on Rockpro64 > > patch 0099: Use LPDDR4-100 on Rock-PI 4 > > Size (increased to ~3KiB ): > - Puma RK3399 (u-boot-spl-dtb.bin): > before: 115644 after: 118744 > - NanoPI M4 (u-boot-tpl-dtb.bin) > before: 41873 after: 44909 > > Travis-CI: > https://travis-ci.org/openedev/u-boot-amarula/builds/546597944 > > Repo: > https://github.com/openedev/u-boot-amarula/tree/rk3399-lpddr4 > > [1] https://patchwork.ozlabs.org/cover/1113893/ > > Any inputs? Was it absolutely necessary to split these changes into 99 commits? I believe at least some of them can be squashed. Reviewing 99 patches isn't feasible. > Jagan. > > Jagan Teki (99): > ram: rk3399: Fix code warnings > ram: rk3399: Add space between string with format specifier > ram: rk3399: Add proper spaces in code > ram: rk3399: s/sdram_params/params > ram: rk3399: Handle data training return types > ram: rk3399: Order include files > ram: rk3399: Move macro after include files > ram: rk3399: Clear PI_175 interrupts in data training > ram: rk3399: Use rank mask in ca data training > ram: rk3399: Use rank mask in wdql data training > ram: rk3399: Add ddrtype enc macro > ram: rk3399: Add channel number encoder macro > ram: rk3399: Add row_3_4 enc macro > ram: rk3399: Add chipinfo macro > ram: rk3399: Add rank enc macro > ram: rk3399: Add column enc macro > ram: rk3399: Add bk enc macro > ram: rk3399: Add dbw enc macro > ram: rk3399: Add cs0_rw macro > ram: rk3399: Add cs1_rw macro > ram: rk3399: Add bw enc macro > ram: rk3399: Rename sys_reg with sys_reg2 > ram: rk3399: Update cs0_row to use sys_reg3 > ram: rk3399: Update cs1_row to use sys_reg3 > ram: rk3399: Add cs1_col enc macro > ram: rk3399: Add ddr version enc macro > ram: rk3399: Add ddrtimingC0 > ram: rk3399: Add DdrMode > ram: rk3399: Handle pctl_cfg return type > ram: rk3399: s/tsel_wr_select_n/tsel_wr_select_dq_n > ram: rk3399: s/tsel_wr_select_p/tsel_wr_select_dq_p > ram: rk3399: s/ca_tsel_wr_select_n/tsel_wr_select_ca_n > ram: rk3399: s/ca_tsel_wr_select_p/tsel_wr_select_ca_p > ram: rk3399: Order tsel variables > ram: rk3399: Add phy pctrl reset support > ram: rk3399: Move pwrup_srefresh_exit to dram_info > ram: rk3399: Add pctl start support > ram: rockchip: rk3399: Add cap_info structure > ram: rk3399: s/rk3399_base_params/sdram_base_params > ram: rk3399: Move common sdram structures in common header > arm: include: rockchip: Move dramtypes to common header > arm: include: rockchip: Add DDR4 enum > ram: rockchip: Add initial Kconfig > debug_uart: Add printdec > ram: rockchip: Add debug sdram driver > ram: rockchip: debug: Add sdram_print_ddr_info > ram: rockchip: debug: Get the cs capacity > ram: rk3399: debug: Add sdram_print_stride > ram: rk3399: Compute stride for 2 channels > ram: rk3399: Compute stride for 1 channel a > ram: rk3399: Add rank detection support > ram: rk3399: Enable sdram debug functions > rockchip: dts: rk3399: nanopi-neo4: Use DDR3-1866 dtsi > clk: rockchip: rk3399: Fix check patch warnings and checks > clk: rockchip: rk3399: Set 50MHz ddr clock > clk: rockchip: rk3399: Set 400MHz ddr clock > ram: rk3399: Add spaces in pctl_cfg > ram: rk3399: Configure phy IO in ds odt > ram: rockchip: Kconfig: Add RK3399 LPDDR4 entry > ram: rk3399: Add lpddr4 rank mask for ca training > ram: rk3399: Add lpddr4 rank mask for wdql training > ram: rk3399: Move mode_sel assignment > ram: rk3399: Don't wait for PLL lock in lpddr4 > ram: rk3399: Avoid two channel ZQ Cal Start at the same time > ram: rk3399: Configure PHY_898, PHY_919 for lpddr4 > ram: rk3399: Configure BOOSTP_EN, BOOSTN_EN for lpddr4 > ram: rk3399: Configure SLEWP_EN, SLEWN_EN for lpddr4 > ram: rk3399: Configure PHY RX_CM_INPUT for lpddr4 > ram: rk3399: Map chipselect for lpddr4 > ram: rk3399: Configure tsel write ca for lpddr4 > ram: rk3399: Don't disable dfi dram clk for lpddr4, rank 1 > ram: rk3399: Add IO settings > ram: sdram: Configure lpddr4 tsel rd, wr based on IO settings > ram: rk3399: Add tsel control clock drive > ram: rk3399: Configure soc odt support > ram: rk3399: Get lpddr4 tsel_rd_en from io settings > ram: rk3399: Update lpddr4 vref based on io settings > ram: rk3399: Update lpddr4 mode_sel based on io settings > ram: rk3399: Update lpddr4 vref_mode_ac > ram: rk3399: Simplify data training first argument > ram: rk3399: Handle data training via ops > ram: rk3399: Add LPPDR4 mr detection > arm: include: rockchip: Add rk3399 pmu file > rockchip: rk3399: syscon: Add pmu support > rockchip: dts: rk3399: Add u-boot,dm-pre-reloc for pmu > ram: rk3399: Add LPPDDR4-400 timings inc > ram: rk3399: Add LPPDDR4-800 timings inc > ram: rk3399: Add set_rate sdram rk3399 ops > ram: rk3399: Add lpddr4 set rate support > ram: rk3399: Set lpddr4 dq odt > ram: rk3399: Set lpddr4 ca odt > ram: rk3399: Set lpddr4 MR3 > ram: rk3399: Set lpddr4 MR12 > ram: rk3399: Set lpddr4 MR14 > configs: rockpro64: Enable LPDDR4 support > configs: rock-pi-4: Enable LPDDR4 support > rockchip: dts: rk3399: Add LPDDR4-100 timings > rockchip: dts: rk3399: rockpro64: Use LPDDR4-100 dtsi > rockchip: dts: rk3399: rock-pi-4: Use LPDDR4-100 dtsi > > arch/arm/dts/rk3399-nanopi-neo4-u-boot.dtsi | 1 + > arch/arm/dts/rk3399-rock-pi-4-u-boot.dtsi | 1 + > arch/arm/dts/rk3399-rockpro64-u-boot.dtsi | 1 + > arch/arm/dts/rk3399-sdram-lpddr4-100.dtsi | 1537 +++++++++++ > arch/arm/dts/rk3399-u-boot.dtsi | 4 + > .../include/asm/arch-rockchip/pmu_rk3399.h | 72 + > arch/arm/include/asm/arch-rockchip/sdram.h | 6 - > .../include/asm/arch-rockchip/sdram_common.h | 90 + > .../include/asm/arch-rockchip/sdram_rk322x.h | 7 - > .../include/asm/arch-rockchip/sdram_rk3399.h | 65 +- > arch/arm/mach-rockchip/rk3399/syscon_rk3399.c | 8 + > configs/rock-pi-4-rk3399_defconfig | 1 + > configs/rockpro64-rk3399_defconfig | 1 + > drivers/clk/rockchip/clk_rk3399.c | 76 +- > drivers/ram/Kconfig | 1 + > drivers/ram/rockchip/Kconfig | 33 + > drivers/ram/rockchip/Makefile | 3 +- > .../ram/rockchip/sdram-rk3399-lpddr4-400.inc | 1570 +++++++++++ > .../ram/rockchip/sdram-rk3399-lpddr4-800.inc | 1570 +++++++++++ > drivers/ram/rockchip/sdram_debug.c | 147 ++ > drivers/ram/rockchip/sdram_rk3399.c | 2289 ++++++++++++++--- > include/debug_uart.h | 19 + > 22 files changed, 7035 insertions(+), 467 deletions(-) > create mode 100644 arch/arm/dts/rk3399-sdram-lpddr4-100.dtsi > create mode 100644 arch/arm/include/asm/arch-rockchip/pmu_rk3399.h > create mode 100644 drivers/ram/rockchip/Kconfig > create mode 100644 drivers/ram/rockchip/sdram-rk3399-lpddr4-400.inc > create mode 100644 drivers/ram/rockchip/sdram-rk3399-lpddr4-800.inc > create mode 100644 drivers/ram/rockchip/sdram_debug.c > > -- > 2.18.0.321.gffc6fa0e3 > > _______________________________________________ > U-Boot mailing list > U-Boot@lists.denx.de > https://lists.denx.de/listinfo/u-boot
> From: Jagan Teki <jagan@amarulasolutions.com> > Date: Mon, 17 Jun 2019 13:01:13 +0530 > > This is the v2 set for supporting LPDDR4 with associated > features, wrt to previous series[1]. > > Thanks to > - YouMin Chen > - Akash Gajjar > - Kever Yang > for supporting all the help on this work. > > On summary this series support > - Code warning and fixes > - rank detection, this would required to probe single channel > sdram configured in NanoPI-NEO4 > - LPDDR4 support, tested in Rockpro64 and Rock-PI-4 > > Changes for v2: > - handle LPDDR4 code as part of CONFIG_RAM_RK3399_LPDDR4 > - support data_training and set_rate via sdram_rk3399_ops > - add proper sys_reg_enc macros > - add new patch to rename variable sdram_params with params > - fix few commit messages > > patch 0001 - 0034: fix code warnings, prints, new macros > > patch 0035 - 0052: rank detection, sdram debug code > > patch 0053: use DDR3-1800 on NanoPI-NEO4 > > patch 0054 - 0094: lpddr4 support > > patch 0095: enable lpddr4 in Rockpro64 > > patch 0096: enable lpddr4 in Rock-PI-4 > > patch 0097: LPDDR4-100 timings > > patch 0098: Use LPDDR4-100 on Rockpro64 > > patch 0099: Use LPDDR4-100 on Rock-PI 4 > > Size (increased to ~3KiB ): > - Puma RK3399 (u-boot-spl-dtb.bin): > before: 115644 after: 118744 > - NanoPI M4 (u-boot-tpl-dtb.bin) > before: 41873 after: 44909 > > Travis-CI: > https://travis-ci.org/openedev/u-boot-amarula/builds/546597944 > > Repo: > https://github.com/openedev/u-boot-amarula/tree/rk3399-lpddr4 > > [1] https://patchwork.ozlabs.org/cover/1113893/ > > Any inputs? > Jagan. > > Jagan Teki (99): > ram: rk3399: Fix code warnings > ram: rk3399: Add space between string with format specifier > ram: rk3399: Add proper spaces in code > ram: rk3399: s/sdram_params/params > ram: rk3399: Handle data training return types > ram: rk3399: Order include files > ram: rk3399: Move macro after include files > ram: rk3399: Clear PI_175 interrupts in data training > ram: rk3399: Use rank mask in ca data training > ram: rk3399: Use rank mask in wdql data training > ram: rk3399: Add ddrtype enc macro > ram: rk3399: Add channel number encoder macro > ram: rk3399: Add row_3_4 enc macro > ram: rk3399: Add chipinfo macro > ram: rk3399: Add rank enc macro > ram: rk3399: Add column enc macro > ram: rk3399: Add bk enc macro > ram: rk3399: Add dbw enc macro > ram: rk3399: Add cs0_rw macro > ram: rk3399: Add cs1_rw macro > ram: rk3399: Add bw enc macro > ram: rk3399: Rename sys_reg with sys_reg2 > ram: rk3399: Update cs0_row to use sys_reg3 > ram: rk3399: Update cs1_row to use sys_reg3 > ram: rk3399: Add cs1_col enc macro > ram: rk3399: Add ddr version enc macro > ram: rk3399: Add ddrtimingC0 > ram: rk3399: Add DdrMode > ram: rk3399: Handle pctl_cfg return type > ram: rk3399: s/tsel_wr_select_n/tsel_wr_select_dq_n > ram: rk3399: s/tsel_wr_select_p/tsel_wr_select_dq_p > ram: rk3399: s/ca_tsel_wr_select_n/tsel_wr_select_ca_n > ram: rk3399: s/ca_tsel_wr_select_p/tsel_wr_select_ca_p > ram: rk3399: Order tsel variables > ram: rk3399: Add phy pctrl reset support > ram: rk3399: Move pwrup_srefresh_exit to dram_info > ram: rk3399: Add pctl start support > ram: rockchip: rk3399: Add cap_info structure > ram: rk3399: s/rk3399_base_params/sdram_base_params > ram: rk3399: Move common sdram structures in common header > arm: include: rockchip: Move dramtypes to common header > arm: include: rockchip: Add DDR4 enum > ram: rockchip: Add initial Kconfig > debug_uart: Add printdec > ram: rockchip: Add debug sdram driver > ram: rockchip: debug: Add sdram_print_ddr_info > ram: rockchip: debug: Get the cs capacity > ram: rk3399: debug: Add sdram_print_stride > ram: rk3399: Compute stride for 2 channels > ram: rk3399: Compute stride for 1 channel a > ram: rk3399: Add rank detection support > ram: rk3399: Enable sdram debug functions > rockchip: dts: rk3399: nanopi-neo4: Use DDR3-1866 dtsi > clk: rockchip: rk3399: Fix check patch warnings and checks > clk: rockchip: rk3399: Set 50MHz ddr clock > clk: rockchip: rk3399: Set 400MHz ddr clock > ram: rk3399: Add spaces in pctl_cfg > ram: rk3399: Configure phy IO in ds odt > ram: rockchip: Kconfig: Add RK3399 LPDDR4 entry > ram: rk3399: Add lpddr4 rank mask for ca training > ram: rk3399: Add lpddr4 rank mask for wdql training > ram: rk3399: Move mode_sel assignment > ram: rk3399: Don't wait for PLL lock in lpddr4 > ram: rk3399: Avoid two channel ZQ Cal Start at the same time > ram: rk3399: Configure PHY_898, PHY_919 for lpddr4 > ram: rk3399: Configure BOOSTP_EN, BOOSTN_EN for lpddr4 > ram: rk3399: Configure SLEWP_EN, SLEWN_EN for lpddr4 > ram: rk3399: Configure PHY RX_CM_INPUT for lpddr4 > ram: rk3399: Map chipselect for lpddr4 > ram: rk3399: Configure tsel write ca for lpddr4 > ram: rk3399: Don't disable dfi dram clk for lpddr4, rank 1 > ram: rk3399: Add IO settings > ram: sdram: Configure lpddr4 tsel rd, wr based on IO settings > ram: rk3399: Add tsel control clock drive > ram: rk3399: Configure soc odt support > ram: rk3399: Get lpddr4 tsel_rd_en from io settings > ram: rk3399: Update lpddr4 vref based on io settings > ram: rk3399: Update lpddr4 mode_sel based on io settings > ram: rk3399: Update lpddr4 vref_mode_ac > ram: rk3399: Simplify data training first argument > ram: rk3399: Handle data training via ops > ram: rk3399: Add LPPDR4 mr detection > arm: include: rockchip: Add rk3399 pmu file > rockchip: rk3399: syscon: Add pmu support > rockchip: dts: rk3399: Add u-boot,dm-pre-reloc for pmu > ram: rk3399: Add LPPDDR4-400 timings inc > ram: rk3399: Add LPPDDR4-800 timings inc > ram: rk3399: Add set_rate sdram rk3399 ops > ram: rk3399: Add lpddr4 set rate support > ram: rk3399: Set lpddr4 dq odt > ram: rk3399: Set lpddr4 ca odt > ram: rk3399: Set lpddr4 MR3 > ram: rk3399: Set lpddr4 MR12 > ram: rk3399: Set lpddr4 MR14 > configs: rockpro64: Enable LPDDR4 support > configs: rock-pi-4: Enable LPDDR4 support > rockchip: dts: rk3399: Add LPDDR4-100 timings > rockchip: dts: rk3399: rockpro64: Use LPDDR4-100 dtsi > rockchip: dts: rk3399: rock-pi-4: Use LPDDR4-100 dtsi > > arch/arm/dts/rk3399-nanopi-neo4-u-boot.dtsi | 1 + > arch/arm/dts/rk3399-rock-pi-4-u-boot.dtsi | 1 + > arch/arm/dts/rk3399-rockpro64-u-boot.dtsi | 1 + > arch/arm/dts/rk3399-sdram-lpddr4-100.dtsi | 1537 +++++++++++ > arch/arm/dts/rk3399-u-boot.dtsi | 4 + > .../include/asm/arch-rockchip/pmu_rk3399.h | 72 + > arch/arm/include/asm/arch-rockchip/sdram.h | 6 - > .../include/asm/arch-rockchip/sdram_common.h | 90 + > .../include/asm/arch-rockchip/sdram_rk322x.h | 7 - > .../include/asm/arch-rockchip/sdram_rk3399.h | 65 +- > arch/arm/mach-rockchip/rk3399/syscon_rk3399.c | 8 + > configs/rock-pi-4-rk3399_defconfig | 1 + > configs/rockpro64-rk3399_defconfig | 1 + > drivers/clk/rockchip/clk_rk3399.c | 76 +- > drivers/ram/Kconfig | 1 + > drivers/ram/rockchip/Kconfig | 33 + > drivers/ram/rockchip/Makefile | 3 +- > .../ram/rockchip/sdram-rk3399-lpddr4-400.inc | 1570 +++++++++++ > .../ram/rockchip/sdram-rk3399-lpddr4-800.inc | 1570 +++++++++++ > drivers/ram/rockchip/sdram_debug.c | 147 ++ > drivers/ram/rockchip/sdram_rk3399.c | 2289 ++++++++++++++--- > include/debug_uart.h | 19 + > 22 files changed, 7035 insertions(+), 467 deletions(-) > create mode 100644 arch/arm/dts/rk3399-sdram-lpddr4-100.dtsi > create mode 100644 arch/arm/include/asm/arch-rockchip/pmu_rk3399.h > create mode 100644 drivers/ram/rockchip/Kconfig > create mode 100644 drivers/ram/rockchip/sdram-rk3399-lpddr4-400.inc > create mode 100644 drivers/ram/rockchip/sdram-rk3399-lpddr4-800.inc > create mode 100644 drivers/ram/rockchip/sdram_debug.c > > -- > 2.18.0.321.gffc6fa0e3 Tested on rockpro64 and firefly-rk3399. Both boards work with this diff provided I apply the two "boot-on" regulator diffs that I posted a couple of days ago. A significant improvement for rockpro64 and no regression for firefly-rk3399. So: Tested-by: Mark Kettenis <kettenis@openbsd.org>
Hi Vasily, On Fri, Jun 21, 2019 at 5:58 AM Vasily Khoruzhick <anarsoul@gmail.com> wrote: > > On Mon, Jun 17, 2019 at 12:37 AM Jagan Teki <jagan@amarulasolutions.com> wrote: > > > > This is the v2 set for supporting LPDDR4 with associated > > features, wrt to previous series[1]. > > > > Thanks to > > - YouMin Chen > > - Akash Gajjar > > - Kever Yang > > for supporting all the help on this work. > > > > On summary this series support > > - Code warning and fixes > > - rank detection, this would required to probe single channel > > sdram configured in NanoPI-NEO4 > > - LPDDR4 support, tested in Rockpro64 and Rock-PI-4 > > > > Changes for v2: > > - handle LPDDR4 code as part of CONFIG_RAM_RK3399_LPDDR4 > > - support data_training and set_rate via sdram_rk3399_ops > > - add proper sys_reg_enc macros > > - add new patch to rename variable sdram_params with params > > - fix few commit messages > > > > patch 0001 - 0034: fix code warnings, prints, new macros > > > > patch 0035 - 0052: rank detection, sdram debug code > > > > patch 0053: use DDR3-1800 on NanoPI-NEO4 > > > > patch 0054 - 0094: lpddr4 support > > > > patch 0095: enable lpddr4 in Rockpro64 > > > > patch 0096: enable lpddr4 in Rock-PI-4 > > > > patch 0097: LPDDR4-100 timings > > > > patch 0098: Use LPDDR4-100 on Rockpro64 > > > > patch 0099: Use LPDDR4-100 on Rock-PI 4 > > > > Size (increased to ~3KiB ): > > - Puma RK3399 (u-boot-spl-dtb.bin): > > before: 115644 after: 118744 > > - NanoPI M4 (u-boot-tpl-dtb.bin) > > before: 41873 after: 44909 > > > > Travis-CI: > > https://travis-ci.org/openedev/u-boot-amarula/builds/546597944 > > > > Repo: > > https://github.com/openedev/u-boot-amarula/tree/rk3399-lpddr4 > > > > [1] https://patchwork.ozlabs.org/cover/1113893/ > > > > Any inputs? > > Was it absolutely necessary to split these changes into 99 commits? I > believe at least some of them can be squashed. Reviewing 99 patches > isn't feasible. Squashed, I'm not sure because the patches were created to satisfy the bisectability and travis-ci, if you find any please feel to comment. About the commit count, I have mentioned in v1, the idea of having many commits in one series to have all lpddr4(-related) changes in one place and also all the commit has incremental approach of supporting rank detection and lpddr4. If require I'm open to sent next versions as multiple series, no problem on that. Jagan.
Hi Jagan, Thanks for your hard work. I'm sure everyone in the Rockchip community is excited about finally having this support in U-Boot. On Tue, 25 Jun 2019 at 12:46, Jagan Teki <jagan@amarulasolutions.com> wrote: [..] > > > > Was it absolutely necessary to split these changes into 99 commits? I > > believe at least some of them can be squashed. Reviewing 99 patches > > isn't feasible. > > Squashed, I'm not sure because the patches were created to satisfy the > bisectability and travis-ci, if you find any please feel to comment. > About the commit count, I have mentioned in v1, the idea of having > many commits in one series to have all lpddr4(-related) changes in one > place and also all the commit has incremental approach of supporting > rank detection and lpddr4. If require I'm open to sent next versions > as multiple series, no problem on that. > I strongly agree with Vasily, and I don't think multiple series makes it any better. What's the reason for having two commits for: "ram: rk3399: Set lpddr4 MR3" and "ram: rk3399: Set lpddr4 MR12" ? Or splitting all the "ram: rk3399: Add ... macro" ? What do you loose if you merge the patches into one? I must confess I am very surprised, and don't really understand what do you gain with this excessive split. Normally, when we are adding a new feature, we normally don't need many patches, so it's the other way around, really. Bisectability is about not breaking existing support, but because the feature is new, normally this is easy. Thanks again! Eze
On Wed, Jun 26, 2019 at 12:12 AM Ezequiel Garcia <ezequiel@vanguardiasur.com.ar> wrote: > > Hi Jagan, > > Thanks for your hard work. I'm sure everyone in the Rockchip community > is excited about finally having this support in U-Boot. > > On Tue, 25 Jun 2019 at 12:46, Jagan Teki <jagan@amarulasolutions.com> wrote: > [..] > > > > > > Was it absolutely necessary to split these changes into 99 commits? I > > > believe at least some of them can be squashed. Reviewing 99 patches > > > isn't feasible. > > > > Squashed, I'm not sure because the patches were created to satisfy the > > bisectability and travis-ci, if you find any please feel to comment. > > About the commit count, I have mentioned in v1, the idea of having > > many commits in one series to have all lpddr4(-related) changes in one > > place and also all the commit has incremental approach of supporting > > rank detection and lpddr4. If require I'm open to sent next versions > > as multiple series, no problem on that. > > > > I strongly agree with Vasily, and I don't think multiple series makes it any > better. > > What's the reason for having two commits for: > > "ram: rk3399: Set lpddr4 MR3" and "ram: rk3399: Set lpddr4 MR12" ? These are individual lpddr4 set rate registers to support, each one is independent on it' own initialization and more over on the whole, it is critical to review. > > Or splitting all the "ram: rk3399: Add ... macro" ? You mean the patches 13 to 20 same like above each one has it's own meaning. It is not meaningful to squash them all. > > What do you loose if you merge the patches into one? > > I must confess I am very surprised, and don't really understand what do you > gain with this excessive split. Normally, when we are adding a new feature, > we normally don't need many patches, so it's the other way around, really. > > Bisectability is about not breaking existing support, but because the feature > is new, normally this is easy. Look, like the whole confusion seems to be because of more patches in one series and the cover-letter states that it support lpddr4. I understand it now, will send the relevant changes in next version accordingly, if require I will squash if any. Jagan.
Jagan, On 06/26/2019 06:22 PM, Jagan Teki wrote: > On Wed, Jun 26, 2019 at 12:12 AM Ezequiel Garcia > <ezequiel@vanguardiasur.com.ar> wrote: >> Hi Jagan, >> >> Thanks for your hard work. I'm sure everyone in the Rockchip community >> is excited about finally having this support in U-Boot. >> >> On Tue, 25 Jun 2019 at 12:46, Jagan Teki <jagan@amarulasolutions.com> wrote: >> [..] >>>> Was it absolutely necessary to split these changes into 99 commits? I >>>> believe at least some of them can be squashed. Reviewing 99 patches >>>> isn't feasible. >>> Squashed, I'm not sure because the patches were created to satisfy the >>> bisectability and travis-ci, if you find any please feel to comment. >>> About the commit count, I have mentioned in v1, the idea of having >>> many commits in one series to have all lpddr4(-related) changes in one >>> place and also all the commit has incremental approach of supporting >>> rank detection and lpddr4. If require I'm open to sent next versions >>> as multiple series, no problem on that. >>> >> I strongly agree with Vasily, and I don't think multiple series makes it any >> better. >> >> What's the reason for having two commits for: >> >> "ram: rk3399: Set lpddr4 MR3" and "ram: rk3399: Set lpddr4 MR12" ? > These are individual lpddr4 set rate registers to support, each one is > independent on it' own initialization and more over on the whole, it > is critical to review. > >> Or splitting all the "ram: rk3399: Add ... macro" ? > You mean the patches 13 to 20 same like above each one has it's own > meaning. It is not meaningful to squash them all. 99 patches is really too much, but I'm not sure how smaller it can be. Reference to kernel document, it suggest not more than 15 at one time: NO!!!! No more huge patch bombs to linux-kernel@vger.kernel.org people! <https://lkml.org/lkml/2005/7/11/336> But please note that I don't think split this into different series make any sense. But maybe you can try to squash as much as you can. eg. the update for dram_all_config may able to squash into one patch, and some new MACRO and its reference code may be able to squash. So it depends on how you define about _logical change_. I'm not sure if this have happen in the history of U-Boot mailing list, but I think this big patch set will be complained by many people if this is send to kernel. I would need to review all the code one by one, and merge them all at one time, so it does not affect much about my review, but for U-Boot patch submit requirement, I'm not sure if any suggestion from other maintainers. Thanks, - Kever >> What do you loose if you merge the patches into one? >> >> I must confess I am very surprised, and don't really understand what do you >> gain with this excessive split. Normally, when we are adding a new feature, >> we normally don't need many patches, so it's the other way around, really. >> >> Bisectability is about not breaking existing support, but because the feature >> is new, normally this is easy. > Look, like the whole confusion seems to be because of more patches in > one series and the cover-letter states that it support lpddr4. I > understand it now, will send the relevant changes in next version > accordingly, if require I will squash if any. > > Jagan. > > _______________________________________________ > Linux-rockchip mailing list > Linux-rockchip@lists.infradead.org > http://lists.infradead.org/mailman/listinfo/linux-rockchip >
Hi Kever, On Thu, Jul 4, 2019 at 3:57 PM Kever Yang <kever.yang@rock-chips.com> wrote: > > Jagan, > > > On 06/26/2019 06:22 PM, Jagan Teki wrote: > > On Wed, Jun 26, 2019 at 12:12 AM Ezequiel Garcia > <ezequiel@vanguardiasur.com.ar> wrote: > > Hi Jagan, > > Thanks for your hard work. I'm sure everyone in the Rockchip community > is excited about finally having this support in U-Boot. > > On Tue, 25 Jun 2019 at 12:46, Jagan Teki <jagan@amarulasolutions.com> wrote: > [..] > > Was it absolutely necessary to split these changes into 99 commits? I > believe at least some of them can be squashed. Reviewing 99 patches > isn't feasible. > > Squashed, I'm not sure because the patches were created to satisfy the > bisectability and travis-ci, if you find any please feel to comment. > About the commit count, I have mentioned in v1, the idea of having > many commits in one series to have all lpddr4(-related) changes in one > place and also all the commit has incremental approach of supporting > rank detection and lpddr4. If require I'm open to sent next versions > as multiple series, no problem on that. > > I strongly agree with Vasily, and I don't think multiple series makes it any > better. > > What's the reason for having two commits for: > > "ram: rk3399: Set lpddr4 MR3" and "ram: rk3399: Set lpddr4 MR12" ? > > These are individual lpddr4 set rate registers to support, each one is > independent on it' own initialization and more over on the whole, it > is critical to review. > > Or splitting all the "ram: rk3399: Add ... macro" ? > > You mean the patches 13 to 20 same like above each one has it's own > meaning. It is not meaningful to squash them all. > > > 99 patches is really too much, but I'm not sure how smaller it can be. > Reference to kernel document, it suggest not more than 15 at one time: Agreed. > NO!!!! No more huge patch bombs to linux-kernel@vger.kernel.org people! <https://lkml.org/lkml/2005/7/11/336> But please note that I don't think split this into different series make any sense. But maybe you can try to squash as much as you can. eg. the update for dram_all_config may able to squash into one patch, > and some new MACRO and its reference code may be able to squash. > So it depends on how you define about _logical change_. > I'm not sure if this have happen in the history of U-Boot mailing list, but > I think this big patch set will be complained by many people if this is send to > kernel. I don't mean to split the lpddr4 changes into multiple series. what I'm trying to say here is this series has patches that support code warnings, rank detection. Since each of them has it own identical features, I'm planning to send them first. and will squash what it require on lpddr4. Will that be okay for you?
Jagan, On 07/04/2019 06:54 PM, Jagan Teki wrote: > Hi Kever, > > On Thu, Jul 4, 2019 at 3:57 PM Kever Yang <kever.yang@rock-chips.com> wrote: >> Jagan, >> >> >> On 06/26/2019 06:22 PM, Jagan Teki wrote: >> >> On Wed, Jun 26, 2019 at 12:12 AM Ezequiel Garcia >> <ezequiel@vanguardiasur.com.ar> wrote: >> >> Hi Jagan, >> >> Thanks for your hard work. I'm sure everyone in the Rockchip community >> is excited about finally having this support in U-Boot. >> >> On Tue, 25 Jun 2019 at 12:46, Jagan Teki <jagan@amarulasolutions.com> wrote: >> [..] >> >> Was it absolutely necessary to split these changes into 99 commits? I >> believe at least some of them can be squashed. Reviewing 99 patches >> isn't feasible. >> >> Squashed, I'm not sure because the patches were created to satisfy the >> bisectability and travis-ci, if you find any please feel to comment. >> About the commit count, I have mentioned in v1, the idea of having >> many commits in one series to have all lpddr4(-related) changes in one >> place and also all the commit has incremental approach of supporting >> rank detection and lpddr4. If require I'm open to sent next versions >> as multiple series, no problem on that. >> >> I strongly agree with Vasily, and I don't think multiple series makes it any >> better. >> >> What's the reason for having two commits for: >> >> "ram: rk3399: Set lpddr4 MR3" and "ram: rk3399: Set lpddr4 MR12" ? >> >> These are individual lpddr4 set rate registers to support, each one is >> independent on it' own initialization and more over on the whole, it >> is critical to review. >> >> Or splitting all the "ram: rk3399: Add ... macro" ? >> >> You mean the patches 13 to 20 same like above each one has it's own >> meaning. It is not meaningful to squash them all. >> >> >> 99 patches is really too much, but I'm not sure how smaller it can be. >> Reference to kernel document, it suggest not more than 15 at one time: > Agreed. > >> NO!!!! No more huge patch bombs to linux-kernel@vger.kernel.org people! <https://lkml.org/lkml/2005/7/11/336> But please note that I don't think split this into different series make any sense. But maybe you can try to squash as much as you can. eg. the update for dram_all_config may able to squash into one patch, >> and some new MACRO and its reference code may be able to squash. >> So it depends on how you define about _logical change_. >> I'm not sure if this have happen in the history of U-Boot mailing list, but >> I think this big patch set will be complained by many people if this is send to >> kernel. > I don't mean to split the lpddr4 changes into multiple series. what > I'm trying to say here is this series has patches that support code > warnings, rank detection. Since each of them has it own identical > features, I'm planning to send them first. and will squash what it > require on lpddr4. > > Will that be okay for you? That's OK for me, just need to let people know the patch set dependency. Thanks, - Kever >