Message ID | 20240906-fix-nt35510-v1-1-1971f3af7dda@linaro.org (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | drm/panel: nt35510: Make new commands optional | expand |
On 9/6/2024 2:54 PM, Linus Walleij wrote: > The commit introducing the Frida display started to write the > SETVCL, BT3CTR and SETVCMOFF registers unconditionally, and some > (not all!) Hydis display seem to be affected by ghosting after > the commit. > > Make them all optional and only send these commands on the > Frida display for now. > > Reported-by: Stefan Hansson <newbyte@postmarketos.org> > Fixes: 219a1f49094f ("drm/panel: nt35510: support FRIDA FRD400B25025-A-CTK") > Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Acked-by: Jessica Zhang <quic_jesszhan@quicinc.com> > --- > Stefan: if this patch works and you have time you can try adding > the three flags one at a time to your device and we can see which > command is problematic. > --- > drivers/gpu/drm/panel/panel-novatek-nt35510.c | 45 +++++++++++++++++---------- > 1 file changed, 29 insertions(+), 16 deletions(-) > > diff --git a/drivers/gpu/drm/panel/panel-novatek-nt35510.c b/drivers/gpu/drm/panel/panel-novatek-nt35510.c > index d3bfdfc9cff6..e07409dcbd1d 100644 > --- a/drivers/gpu/drm/panel/panel-novatek-nt35510.c > +++ b/drivers/gpu/drm/panel/panel-novatek-nt35510.c > @@ -38,6 +38,9 @@ > > #define NT35510_CMD_CORRECT_GAMMA BIT(0) > #define NT35510_CMD_CONTROL_DISPLAY BIT(1) > +#define NT35510_CMD_SETVCL BIT(2) > +#define NT35510_CMD_BT3CTR BIT(3) > +#define NT35510_CMD_SETVCMOFF BIT(4) > > #define MCS_CMD_MAUCCTR 0xF0 /* Manufacturer command enable */ > #define MCS_CMD_READ_ID1 0xDA > @@ -675,16 +678,23 @@ static int nt35510_setup_power(struct nt35510 *nt) > nt->conf->bt2ctr); > if (ret) > return ret; > - ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCL, > - NT35510_P1_VCL_LEN, > - nt->conf->vcl); > - if (ret) > - return ret; > - ret = nt35510_send_long(nt, dsi, NT35510_P1_BT3CTR, > - NT35510_P1_BT3CTR_LEN, > - nt->conf->bt3ctr); > - if (ret) > - return ret; > + > + if (nt->conf->cmds & NT35510_CMD_SETVCL) { > + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCL, > + NT35510_P1_VCL_LEN, > + nt->conf->vcl); > + if (ret) > + return ret; > + } > + > + if (nt->conf->cmds & NT35510_CMD_BT3CTR) { > + ret = nt35510_send_long(nt, dsi, NT35510_P1_BT3CTR, > + NT35510_P1_BT3CTR_LEN, > + nt->conf->bt3ctr); > + if (ret) > + return ret; > + } > + > ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVGH, > NT35510_P1_VGH_LEN, > nt->conf->vgh); > @@ -721,11 +731,13 @@ static int nt35510_setup_power(struct nt35510 *nt) > if (ret) > return ret; > > - ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCMOFF, > - NT35510_P1_VCMOFF_LEN, > - nt->conf->vcmoff); > - if (ret) > - return ret; > + if (nt->conf->cmds & NT35510_CMD_SETVCMOFF) { > + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCMOFF, > + NT35510_P1_VCMOFF_LEN, > + nt->conf->vcmoff); > + if (ret) > + return ret; > + } > > /* Typically 10 ms */ > usleep_range(10000, 20000); > @@ -1319,7 +1331,8 @@ static const struct nt35510_config nt35510_frida_frd400b25025 = { > }, > .mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST | > MIPI_DSI_MODE_LPM, > - .cmds = NT35510_CMD_CONTROL_DISPLAY, > + .cmds = NT35510_CMD_CONTROL_DISPLAY | NT35510_CMD_SETVCL | > + NT35510_CMD_BT3CTR | NT35510_CMD_SETVCMOFF, > /* 0x03: AVDD = 6.2V */ > .avdd = { 0x03, 0x03, 0x03 }, > /* 0x46: PCK = 2 x Hsync, BTP = 2.5 x VDDB */ > > --- > base-commit: 8400291e289ee6b2bf9779ff1c83a291501f017b > change-id: 20240906-fix-nt35510-a8ec6e47e036 > > Best regards, > -- > Linus Walleij <linus.walleij@linaro.org> >
Hi Linus! On 2024-09-06 23:54, Linus Walleij wrote: > The commit introducing the Frida display started to write the > SETVCL, BT3CTR and SETVCMOFF registers unconditionally, and some > (not all!) Hydis display seem to be affected by ghosting after > the commit. > > Make them all optional and only send these commands on the > Frida display for now. > > Reported-by: Stefan Hansson <newbyte@postmarketos.org> > Fixes: 219a1f49094f ("drm/panel: nt35510: support FRIDA FRD400B25025-A-CTK") > Signed-off-by: Linus Walleij <linus.walleij@linaro.org> > --- > Stefan: if this patch works and you have time you can try adding > the three flags one at a time to your device and we can see which > command is problematic. I did just that. I tried different combinations of the .cmds property of nt35510_hydis_hva40wv1 and came to these conclusions: Good: .cmds = NT35510_CMD_CORRECT_GAMMA .cmds = NT35510_CMD_CORRECT_GAMMA | NT35510_CMD_SETVCL .cmds = NT35510_CMD_CORRECT_GAMMA | NT35510_CMD_SETVCL | NT35510_CMD_BT3CTR Bad: .cmds = NT35510_CMD_CORRECT_GAMMA | NT35510_CMD_SETVCL | NT35510_CMD_BT3CTR | NT35510_CMD_SETVCMOFF .cmds = NT35510_CMD_CORRECT_GAMMA | NT35510_CMD_SETVCMOFF By good, I mean "no visible ghosting" and by bad I mean "visible ghosting". Based on this, it seems to me that NT35510_CMD_SETVCMOFF is the culprit. However, this patch is also good as-is. > --- > drivers/gpu/drm/panel/panel-novatek-nt35510.c | 45 +++++++++++++++++---------- > 1 file changed, 29 insertions(+), 16 deletions(-) > > diff --git a/drivers/gpu/drm/panel/panel-novatek-nt35510.c b/drivers/gpu/drm/panel/panel-novatek-nt35510.c > index d3bfdfc9cff6..e07409dcbd1d 100644 > --- a/drivers/gpu/drm/panel/panel-novatek-nt35510.c > +++ b/drivers/gpu/drm/panel/panel-novatek-nt35510.c > @@ -38,6 +38,9 @@ > > #define NT35510_CMD_CORRECT_GAMMA BIT(0) > #define NT35510_CMD_CONTROL_DISPLAY BIT(1) > +#define NT35510_CMD_SETVCL BIT(2) > +#define NT35510_CMD_BT3CTR BIT(3) > +#define NT35510_CMD_SETVCMOFF BIT(4) > > #define MCS_CMD_MAUCCTR 0xF0 /* Manufacturer command enable */ > #define MCS_CMD_READ_ID1 0xDA > @@ -675,16 +678,23 @@ static int nt35510_setup_power(struct nt35510 *nt) > nt->conf->bt2ctr); > if (ret) > return ret; > - ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCL, > - NT35510_P1_VCL_LEN, > - nt->conf->vcl); > - if (ret) > - return ret; > - ret = nt35510_send_long(nt, dsi, NT35510_P1_BT3CTR, > - NT35510_P1_BT3CTR_LEN, > - nt->conf->bt3ctr); > - if (ret) > - return ret; > + > + if (nt->conf->cmds & NT35510_CMD_SETVCL) { > + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCL, > + NT35510_P1_VCL_LEN, > + nt->conf->vcl); > + if (ret) > + return ret; > + } > + > + if (nt->conf->cmds & NT35510_CMD_BT3CTR) { > + ret = nt35510_send_long(nt, dsi, NT35510_P1_BT3CTR, > + NT35510_P1_BT3CTR_LEN, > + nt->conf->bt3ctr); > + if (ret) > + return ret; > + } > + > ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVGH, > NT35510_P1_VGH_LEN, > nt->conf->vgh); > @@ -721,11 +731,13 @@ static int nt35510_setup_power(struct nt35510 *nt) > if (ret) > return ret; > > - ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCMOFF, > - NT35510_P1_VCMOFF_LEN, > - nt->conf->vcmoff); > - if (ret) > - return ret; > + if (nt->conf->cmds & NT35510_CMD_SETVCMOFF) { > + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCMOFF, > + NT35510_P1_VCMOFF_LEN, > + nt->conf->vcmoff); > + if (ret) > + return ret; > + } > > /* Typically 10 ms */ > usleep_range(10000, 20000); > @@ -1319,7 +1331,8 @@ static const struct nt35510_config nt35510_frida_frd400b25025 = { > }, > .mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST | > MIPI_DSI_MODE_LPM, > - .cmds = NT35510_CMD_CONTROL_DISPLAY, > + .cmds = NT35510_CMD_CONTROL_DISPLAY | NT35510_CMD_SETVCL | > + NT35510_CMD_BT3CTR | NT35510_CMD_SETVCMOFF, > /* 0x03: AVDD = 6.2V */ > .avdd = { 0x03, 0x03, 0x03 }, > /* 0x46: PCK = 2 x Hsync, BTP = 2.5 x VDDB */ > > --- > base-commit: 8400291e289ee6b2bf9779ff1c83a291501f017b > change-id: 20240906-fix-nt35510-a8ec6e47e036 > > Best regards, Stefan Hansson
diff --git a/drivers/gpu/drm/panel/panel-novatek-nt35510.c b/drivers/gpu/drm/panel/panel-novatek-nt35510.c index d3bfdfc9cff6..e07409dcbd1d 100644 --- a/drivers/gpu/drm/panel/panel-novatek-nt35510.c +++ b/drivers/gpu/drm/panel/panel-novatek-nt35510.c @@ -38,6 +38,9 @@ #define NT35510_CMD_CORRECT_GAMMA BIT(0) #define NT35510_CMD_CONTROL_DISPLAY BIT(1) +#define NT35510_CMD_SETVCL BIT(2) +#define NT35510_CMD_BT3CTR BIT(3) +#define NT35510_CMD_SETVCMOFF BIT(4) #define MCS_CMD_MAUCCTR 0xF0 /* Manufacturer command enable */ #define MCS_CMD_READ_ID1 0xDA @@ -675,16 +678,23 @@ static int nt35510_setup_power(struct nt35510 *nt) nt->conf->bt2ctr); if (ret) return ret; - ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCL, - NT35510_P1_VCL_LEN, - nt->conf->vcl); - if (ret) - return ret; - ret = nt35510_send_long(nt, dsi, NT35510_P1_BT3CTR, - NT35510_P1_BT3CTR_LEN, - nt->conf->bt3ctr); - if (ret) - return ret; + + if (nt->conf->cmds & NT35510_CMD_SETVCL) { + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCL, + NT35510_P1_VCL_LEN, + nt->conf->vcl); + if (ret) + return ret; + } + + if (nt->conf->cmds & NT35510_CMD_BT3CTR) { + ret = nt35510_send_long(nt, dsi, NT35510_P1_BT3CTR, + NT35510_P1_BT3CTR_LEN, + nt->conf->bt3ctr); + if (ret) + return ret; + } + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVGH, NT35510_P1_VGH_LEN, nt->conf->vgh); @@ -721,11 +731,13 @@ static int nt35510_setup_power(struct nt35510 *nt) if (ret) return ret; - ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCMOFF, - NT35510_P1_VCMOFF_LEN, - nt->conf->vcmoff); - if (ret) - return ret; + if (nt->conf->cmds & NT35510_CMD_SETVCMOFF) { + ret = nt35510_send_long(nt, dsi, NT35510_P1_SETVCMOFF, + NT35510_P1_VCMOFF_LEN, + nt->conf->vcmoff); + if (ret) + return ret; + } /* Typically 10 ms */ usleep_range(10000, 20000); @@ -1319,7 +1331,8 @@ static const struct nt35510_config nt35510_frida_frd400b25025 = { }, .mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST | MIPI_DSI_MODE_LPM, - .cmds = NT35510_CMD_CONTROL_DISPLAY, + .cmds = NT35510_CMD_CONTROL_DISPLAY | NT35510_CMD_SETVCL | + NT35510_CMD_BT3CTR | NT35510_CMD_SETVCMOFF, /* 0x03: AVDD = 6.2V */ .avdd = { 0x03, 0x03, 0x03 }, /* 0x46: PCK = 2 x Hsync, BTP = 2.5 x VDDB */
The commit introducing the Frida display started to write the SETVCL, BT3CTR and SETVCMOFF registers unconditionally, and some (not all!) Hydis display seem to be affected by ghosting after the commit. Make them all optional and only send these commands on the Frida display for now. Reported-by: Stefan Hansson <newbyte@postmarketos.org> Fixes: 219a1f49094f ("drm/panel: nt35510: support FRIDA FRD400B25025-A-CTK") Signed-off-by: Linus Walleij <linus.walleij@linaro.org> --- Stefan: if this patch works and you have time you can try adding the three flags one at a time to your device and we can see which command is problematic. --- drivers/gpu/drm/panel/panel-novatek-nt35510.c | 45 +++++++++++++++++---------- 1 file changed, 29 insertions(+), 16 deletions(-) --- base-commit: 8400291e289ee6b2bf9779ff1c83a291501f017b change-id: 20240906-fix-nt35510-a8ec6e47e036 Best regards,