According to the hardware manual [1], the DMAC found in SH7709 features
only 4 channels.
While at it, also sort the existing targets and clarify that
NR_ONCHIP_DMA_CHANNELS must be a multiply of two.
[1] https://www.renesas.com/us/en/document/mah/sh7709s-group-hardware-manual (p. 373)
Signed-off-by: Artur Rojek <contact@artur-rojek.eu>
---
v2: - sort existing targets
- clarify that the value must be a multiply of two
arch/sh/drivers/dma/Kconfig | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/arch/sh/drivers/dma/Kconfig b/arch/sh/drivers/dma/Kconfig
index 7d54f284ce10..382fbb189fcf 100644
--- a/arch/sh/drivers/dma/Kconfig
+++ b/arch/sh/drivers/dma/Kconfig
@@ -28,17 +28,19 @@ config SH_DMA_API
config NR_ONCHIP_DMA_CHANNELS
int
depends on SH_DMA
- default "4" if CPU_SUBTYPE_SH7750 || CPU_SUBTYPE_SH7751 || \
- CPU_SUBTYPE_SH7750S || CPU_SUBTYPE_SH7091
+ default "4" if CPU_SUBTYPE_SH7709 || CPU_SUBTYPE_SH7750 || \
+ CPU_SUBTYPE_SH7750S || CPU_SUBTYPE_SH7751 || \
+ CPU_SUBTYPE_SH7091
default "8" if CPU_SUBTYPE_SH7750R || CPU_SUBTYPE_SH7751R || \
CPU_SUBTYPE_SH7760
- default "12" if CPU_SUBTYPE_SH7723 || CPU_SUBTYPE_SH7780 || \
- CPU_SUBTYPE_SH7785 || CPU_SUBTYPE_SH7724
+ default "12" if CPU_SUBTYPE_SH7723 || CPU_SUBTYPE_SH7724 || \
+ CPU_SUBTYPE_SH7780 || CPU_SUBTYPE_SH7785
default "6"
help
This allows you to specify the number of channels that the on-chip
- DMAC supports. This will be 4 for SH7750/SH7751/Sh7750S/SH7091 and 8 for the
- SH7750R/SH7751R/SH7760, 12 for the SH7723/SH7780/SH7785/SH7724, default is 6.
+ DMAC supports. This will be 4 for SH7709/SH7750/SH7750S/SH7751/SH7091,
+ 8 for SH7750R/SH7751R/SH7760, and 12 for SH7723/SH7724/SH7780/SH7785.
+ Default is 6. Must be an even number.
config SH_DMABRG
bool "SH7760 DMABRG support"
--
2.40.1On Sat, 2023-05-27 at 18:44 +0200, Artur Rojek wrote: > According to the hardware manual [1], the DMAC found in SH7709 features > only 4 channels. > > While at it, also sort the existing targets and clarify that > NR_ONCHIP_DMA_CHANNELS must be a multiply of two. > > [1] https://www.renesas.com/us/en/document/mah/sh7709s-group-hardware-manual (p. 373) > > Signed-off-by: Artur Rojek <contact@artur-rojek.eu> > --- > > v2: - sort existing targets > - clarify that the value must be a multiply of two > > arch/sh/drivers/dma/Kconfig | 14 ++++++++------ > 1 file changed, 8 insertions(+), 6 deletions(-) > > diff --git a/arch/sh/drivers/dma/Kconfig b/arch/sh/drivers/dma/Kconfig > index 7d54f284ce10..382fbb189fcf 100644 > --- a/arch/sh/drivers/dma/Kconfig > +++ b/arch/sh/drivers/dma/Kconfig > @@ -28,17 +28,19 @@ config SH_DMA_API > config NR_ONCHIP_DMA_CHANNELS > int > depends on SH_DMA > - default "4" if CPU_SUBTYPE_SH7750 || CPU_SUBTYPE_SH7751 || \ > - CPU_SUBTYPE_SH7750S || CPU_SUBTYPE_SH7091 > + default "4" if CPU_SUBTYPE_SH7709 || CPU_SUBTYPE_SH7750 || \ > + CPU_SUBTYPE_SH7750S || CPU_SUBTYPE_SH7751 || \ > + CPU_SUBTYPE_SH7091 > default "8" if CPU_SUBTYPE_SH7750R || CPU_SUBTYPE_SH7751R || \ > CPU_SUBTYPE_SH7760 > - default "12" if CPU_SUBTYPE_SH7723 || CPU_SUBTYPE_SH7780 || \ > - CPU_SUBTYPE_SH7785 || CPU_SUBTYPE_SH7724 > + default "12" if CPU_SUBTYPE_SH7723 || CPU_SUBTYPE_SH7724 || \ > + CPU_SUBTYPE_SH7780 || CPU_SUBTYPE_SH7785 > default "6" > help > This allows you to specify the number of channels that the on-chip > - DMAC supports. This will be 4 for SH7750/SH7751/Sh7750S/SH7091 and 8 for the > - SH7750R/SH7751R/SH7760, 12 for the SH7723/SH7780/SH7785/SH7724, default is 6. > + DMAC supports. This will be 4 for SH7709/SH7750/SH7750S/SH7751/SH7091, > + 8 for SH7750R/SH7751R/SH7760, and 12 for SH7723/SH7724/SH7780/SH7785. > + Default is 6. Must be an even number. > > config SH_DMABRG > bool "SH7760 DMABRG support" Reviewed-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer `. `' Physicist `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Hi Artur,
On Sat, May 27, 2023 at 6:45 PM Artur Rojek <contact@artur-rojek.eu> wrote:
> According to the hardware manual [1], the DMAC found in SH7709 features
> only 4 channels.
>
> While at it, also sort the existing targets and clarify that
> NR_ONCHIP_DMA_CHANNELS must be a multiply of two.
>
> [1] https://www.renesas.com/us/en/document/mah/sh7709s-group-hardware-manual (p. 373)
>
> Signed-off-by: Artur Rojek <contact@artur-rojek.eu>
> ---
>
> v2: - sort existing targets
Thanks for the update!
> - clarify that the value must be a multiply of two
That's only true when there are two DMACs, right?
Even in that case, you could mitigate that by avoiding the division by
#ifdef SH_DMAC_BASE1
-#define SH_DMAC_NR_MD_CH (CONFIG_NR_ONCHIP_DMA_CHANNELS / 2)
+#define SH_DMAC_NR_MD_CH 6
#else
#define SH_DMAC_NR_MD_CH CONFIG_NR_ONCHIP_DMA_CHANNELS
#endif
That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS
when configuring his kernel, thus breaking DMA due to an incorrect
value of SH_DMAC_NR_MD_CH.
Unfortunately we cannot protect against that when using a single DMAC,
as SH_DMAC_NR_MD_CH can be either 4, 6, or 8.
Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>,
to protect against a user overriding this value?
> --- a/arch/sh/drivers/dma/Kconfig
> +++ b/arch/sh/drivers/dma/Kconfig
> @@ -28,17 +28,19 @@ config SH_DMA_API
> config NR_ONCHIP_DMA_CHANNELS
> int
> depends on SH_DMA
> - default "4" if CPU_SUBTYPE_SH7750 || CPU_SUBTYPE_SH7751 || \
> - CPU_SUBTYPE_SH7750S || CPU_SUBTYPE_SH7091
> + default "4" if CPU_SUBTYPE_SH7709 || CPU_SUBTYPE_SH7750 || \
> + CPU_SUBTYPE_SH7750S || CPU_SUBTYPE_SH7751 || \
> + CPU_SUBTYPE_SH7091
> default "8" if CPU_SUBTYPE_SH7750R || CPU_SUBTYPE_SH7751R || \
> CPU_SUBTYPE_SH7760
> - default "12" if CPU_SUBTYPE_SH7723 || CPU_SUBTYPE_SH7780 || \
> - CPU_SUBTYPE_SH7785 || CPU_SUBTYPE_SH7724
> + default "12" if CPU_SUBTYPE_SH7723 || CPU_SUBTYPE_SH7724 || \
> + CPU_SUBTYPE_SH7780 || CPU_SUBTYPE_SH7785
> default "6"
> help
> This allows you to specify the number of channels that the on-chip
> - DMAC supports. This will be 4 for SH7750/SH7751/Sh7750S/SH7091 and 8 for the
> - SH7750R/SH7751R/SH7760, 12 for the SH7723/SH7780/SH7785/SH7724, default is 6.
> + DMAC supports. This will be 4 for SH7709/SH7750/SH7750S/SH7751/SH7091,
> + 8 for SH7750R/SH7751R/SH7760, and 12 for SH7723/SH7724/SH7780/SH7785.
> + Default is 6. Must be an even number.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi! Sorry for being so late to the party. On Wed, 2023-06-07 at 11:16 +0200, Geert Uytterhoeven wrote: > Hi Artur, > > On Sat, May 27, 2023 at 6:45 PM Artur Rojek <contact@artur-rojek.eu> wrote: > > According to the hardware manual [1], the DMAC found in SH7709 features > > only 4 channels. > > > > While at it, also sort the existing targets and clarify that > > NR_ONCHIP_DMA_CHANNELS must be a multiply of two. > > > > [1] https://www.renesas.com/us/en/document/mah/sh7709s-group-hardware-manual (p. 373) > > > > Signed-off-by: Artur Rojek <contact@artur-rojek.eu> > > --- > > > > v2: - sort existing targets > > Thanks for the update! > > > - clarify that the value must be a multiply of two > > That's only true when there are two DMACs, right? > > Even in that case, you could mitigate that by avoiding the division by > > #ifdef SH_DMAC_BASE1 > -#define SH_DMAC_NR_MD_CH (CONFIG_NR_ONCHIP_DMA_CHANNELS / 2) > +#define SH_DMAC_NR_MD_CH 6 > #else > #define SH_DMAC_NR_MD_CH CONFIG_NR_ONCHIP_DMA_CHANNELS > #endif Aren't we dropping SH_DMAC_BASE1 in the other patch anyway? > That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS > when configuring his kernel, thus breaking DMA due to an incorrect > value of SH_DMAC_NR_MD_CH. > > Unfortunately we cannot protect against that when using a single DMAC, > as SH_DMAC_NR_MD_CH can be either 4, 6, or 8. > > Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>, > to protect against a user overriding this value? Isn't SH_DMAC_NR_MD_CH already hardwired to the SoC being used? Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer `. `' Physicist `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Hi Adrian,
On Thu, Jun 8, 2023 at 11:54 AM John Paul Adrian Glaubitz
<glaubitz@physik.fu-berlin.de> wrote:
> On Wed, 2023-06-07 at 11:16 +0200, Geert Uytterhoeven wrote:
> > On Sat, May 27, 2023 at 6:45 PM Artur Rojek <contact@artur-rojek.eu> wrote:
> > > According to the hardware manual [1], the DMAC found in SH7709 features
> > > only 4 channels.
> > >
> > > While at it, also sort the existing targets and clarify that
> > > NR_ONCHIP_DMA_CHANNELS must be a multiply of two.
> > >
> > > [1] https://www.renesas.com/us/en/document/mah/sh7709s-group-hardware-manual (p. 373)
> > >
> > > Signed-off-by: Artur Rojek <contact@artur-rojek.eu>
> > > ---
> > >
> > > v2: - sort existing targets
> >
> > Thanks for the update!
> >
> > > - clarify that the value must be a multiply of two
> >
> > That's only true when there are two DMACs, right?
> >
> > Even in that case, you could mitigate that by avoiding the division by
> >
> > #ifdef SH_DMAC_BASE1
> > -#define SH_DMAC_NR_MD_CH (CONFIG_NR_ONCHIP_DMA_CHANNELS / 2)
> > +#define SH_DMAC_NR_MD_CH 6
> > #else
> > #define SH_DMAC_NR_MD_CH CONFIG_NR_ONCHIP_DMA_CHANNELS
> > #endif
>
> Aren't we dropping SH_DMAC_BASE1 in the other patch anyway?
Only for the SH4 parts that do not have it.
It is still set in arch/sh/include/cpu-sh4a/cpu/dma.h for the SH4a parts with
12 channels and 2 DMACs.
> > That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS
> > when configuring his kernel, thus breaking DMA due to an incorrect
> > value of SH_DMAC_NR_MD_CH.
> >
> > Unfortunately we cannot protect against that when using a single DMAC,
> > as SH_DMAC_NR_MD_CH can be either 4, 6, or 8.
> >
> > Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>,
> > to protect against a user overriding this value?
>
> Isn't SH_DMAC_NR_MD_CH already hardwired to the SoC being used?
It depends on CONFIG_NR_ONCHIP_DMA_CHANNELS, while it
should be fixed based on the SoC.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi Geert! On Thu, 2023-06-08 at 11:58 +0200, Geert Uytterhoeven wrote: > > > > Aren't we dropping SH_DMAC_BASE1 in the other patch anyway? > > Only for the SH4 parts that do not have it. > It is still set in arch/sh/include/cpu-sh4a/cpu/dma.h for the SH4a parts with > 12 channels and 2 DMACs. OK, thanks for the clarification. I will review the other patches tonight. > > > That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS > > > when configuring his kernel, thus breaking DMA due to an incorrect > > > value of SH_DMAC_NR_MD_CH. > > > > > > Unfortunately we cannot protect against that when using a single DMAC, > > > as SH_DMAC_NR_MD_CH can be either 4, 6, or 8. > > > > > > Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>, > > > to protect against a user overriding this value? > > > > Isn't SH_DMAC_NR_MD_CH already hardwired to the SoC being used? > > It depends on CONFIG_NR_ONCHIP_DMA_CHANNELS, while it > should be fixed based on the SoC. I agree. However, I would be fine with merging this patch set first and fixing this particular issue in a follow-up series. Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer `. `' Physicist `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Hi Geert! On Thu, 2023-06-08 at 12:03 +0200, John Paul Adrian Glaubitz wrote: > > > > That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS > > > > when configuring his kernel, thus breaking DMA due to an incorrect > > > > value of SH_DMAC_NR_MD_CH. > > > > > > > > Unfortunately we cannot protect against that when using a single DMAC, > > > > as SH_DMAC_NR_MD_CH can be either 4, 6, or 8. > > > > > > > > Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>, > > > > to protect against a user overriding this value? > > > > > > Isn't SH_DMAC_NR_MD_CH already hardwired to the SoC being used? > > > > It depends on CONFIG_NR_ONCHIP_DMA_CHANNELS, while it > > should be fixed based on the SoC. > > I agree. However, I would be fine with merging this patch set first and fixing > this particular issue in a follow-up series. So, my suggestion is to take this series as-is for 6.5, then get the other issues you mentioned fixed for 6.6. I think it's already a gain when these issues are fixed and the kernel boots on the HP Journada 680 again. Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer `. `' Physicist `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Hi Adrian,
On Sat, Jun 17, 2023 at 9:32 AM John Paul Adrian Glaubitz
<glaubitz@physik.fu-berlin.de> wrote:
> On Thu, 2023-06-08 at 12:03 +0200, John Paul Adrian Glaubitz wrote:
> > > > > That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS
> > > > > when configuring his kernel, thus breaking DMA due to an incorrect
> > > > > value of SH_DMAC_NR_MD_CH.
> > > > >
> > > > > Unfortunately we cannot protect against that when using a single DMAC,
> > > > > as SH_DMAC_NR_MD_CH can be either 4, 6, or 8.
> > > > >
> > > > > Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>,
> > > > > to protect against a user overriding this value?
> > > >
> > > > Isn't SH_DMAC_NR_MD_CH already hardwired to the SoC being used?
> > >
> > > It depends on CONFIG_NR_ONCHIP_DMA_CHANNELS, while it
> > > should be fixed based on the SoC.
> >
> > I agree. However, I would be fine with merging this patch set first and fixing
> > this particular issue in a follow-up series.
>
> So, my suggestion is to take this series as-is for 6.5, then get the other issues
> you mentioned fixed for 6.6. I think it's already a gain when these issues are
> fixed and the kernel boots on the HP Journada 680 again.
Sure, I don't want to block the acceptance of this series at all.
Thanks!
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi Geert! On Sat, 2023-06-17 at 13:09 +0200, Geert Uytterhoeven wrote: > Hi Adrian, > > On Sat, Jun 17, 2023 at 9:32 AM John Paul Adrian Glaubitz > <glaubitz@physik.fu-berlin.de> wrote: > > On Thu, 2023-06-08 at 12:03 +0200, John Paul Adrian Glaubitz wrote: > > > > > > That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS > > > > > > when configuring his kernel, thus breaking DMA due to an incorrect > > > > > > value of SH_DMAC_NR_MD_CH. > > > > > > > > > > > > Unfortunately we cannot protect against that when using a single DMAC, > > > > > > as SH_DMAC_NR_MD_CH can be either 4, 6, or 8. > > > > > > > > > > > > Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>, > > > > > > to protect against a user overriding this value? > > > > > > > > > > Isn't SH_DMAC_NR_MD_CH already hardwired to the SoC being used? > > > > > > > > It depends on CONFIG_NR_ONCHIP_DMA_CHANNELS, while it > > > > should be fixed based on the SoC. > > > > > > I agree. However, I would be fine with merging this patch set first and fixing > > > this particular issue in a follow-up series. > > > > So, my suggestion is to take this series as-is for 6.5, then get the other issues > > you mentioned fixed for 6.6. I think it's already a gain when these issues are > > fixed and the kernel boots on the HP Journada 680 again. > > Sure, I don't want to block the acceptance of this series at all. > Thanks! Apologies for the late reply. Would you mind adding your Reviewed-by to this patch before I review and apply the series? Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer `. `' Physicist `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Hi Adrian,
On Tue, Jul 4, 2023 at 7:45 AM John Paul Adrian Glaubitz
<glaubitz@physik.fu-berlin.de> wrote:
> On Sat, 2023-06-17 at 13:09 +0200, Geert Uytterhoeven wrote:
> > On Sat, Jun 17, 2023 at 9:32 AM John Paul Adrian Glaubitz
> > <glaubitz@physik.fu-berlin.de> wrote:
> > > On Thu, 2023-06-08 at 12:03 +0200, John Paul Adrian Glaubitz wrote:
> > > > > > > That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS
> > > > > > > when configuring his kernel, thus breaking DMA due to an incorrect
> > > > > > > value of SH_DMAC_NR_MD_CH.
> > > > > > >
> > > > > > > Unfortunately we cannot protect against that when using a single DMAC,
> > > > > > > as SH_DMAC_NR_MD_CH can be either 4, 6, or 8.
> > > > > > >
> > > > > > > Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>,
> > > > > > > to protect against a user overriding this value?
> > > > > >
> > > > > > Isn't SH_DMAC_NR_MD_CH already hardwired to the SoC being used?
> > > > >
> > > > > It depends on CONFIG_NR_ONCHIP_DMA_CHANNELS, while it
> > > > > should be fixed based on the SoC.
> > > >
> > > > I agree. However, I would be fine with merging this patch set first and fixing
> > > > this particular issue in a follow-up series.
> > >
> > > So, my suggestion is to take this series as-is for 6.5, then get the other issues
> > > you mentioned fixed for 6.6. I think it's already a gain when these issues are
> > > fixed and the kernel boots on the HP Journada 680 again.
> >
> > Sure, I don't want to block the acceptance of this series at all.
> > Thanks!
>
> Apologies for the late reply. Would you mind adding your Reviewed-by to this patch
> before I review and apply the series?
With "must be a multiply of two" and "Must be an even number" removed.
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi Geert! On Tue, 2023-07-04 at 09:32 +0200, Geert Uytterhoeven wrote: > Hi Adrian, > > On Tue, Jul 4, 2023 at 7:45 AM John Paul Adrian Glaubitz > <glaubitz@physik.fu-berlin.de> wrote: > > On Sat, 2023-06-17 at 13:09 +0200, Geert Uytterhoeven wrote: > > > On Sat, Jun 17, 2023 at 9:32 AM John Paul Adrian Glaubitz > > > <glaubitz@physik.fu-berlin.de> wrote: > > > > On Thu, 2023-06-08 at 12:03 +0200, John Paul Adrian Glaubitz wrote: > > > > > > > > That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS > > > > > > > > when configuring his kernel, thus breaking DMA due to an incorrect > > > > > > > > value of SH_DMAC_NR_MD_CH. > > > > > > > > > > > > > > > > Unfortunately we cannot protect against that when using a single DMAC, > > > > > > > > as SH_DMAC_NR_MD_CH can be either 4, 6, or 8. > > > > > > > > > > > > > > > > Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>, > > > > > > > > to protect against a user overriding this value? > > > > > > > > > > > > > > Isn't SH_DMAC_NR_MD_CH already hardwired to the SoC being used? > > > > > > > > > > > > It depends on CONFIG_NR_ONCHIP_DMA_CHANNELS, while it > > > > > > should be fixed based on the SoC. > > > > > > > > > > I agree. However, I would be fine with merging this patch set first and fixing > > > > > this particular issue in a follow-up series. > > > > > > > > So, my suggestion is to take this series as-is for 6.5, then get the other issues > > > > you mentioned fixed for 6.6. I think it's already a gain when these issues are > > > > fixed and the kernel boots on the HP Journada 680 again. > > > > > > Sure, I don't want to block the acceptance of this series at all. > > > Thanks! > > > > Apologies for the late reply. Would you mind adding your Reviewed-by to this patch > > before I review and apply the series? > > With "must be a multiply of two" and "Must be an even number" removed. > Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be> Thanks. I guess, I will drop the whole "and clarify that NR_ONCHIP_DMA_CHANNELS must be a multiply of two" then. Correct? Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer `. `' Physicist `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Hi Adrian.
On Tue, Jul 4, 2023 at 9:43 AM John Paul Adrian Glaubitz
<glaubitz@physik.fu-berlin.de> wrote:
> On Tue, 2023-07-04 at 09:32 +0200, Geert Uytterhoeven wrote:
> > On Tue, Jul 4, 2023 at 7:45 AM John Paul Adrian Glaubitz
> > <glaubitz@physik.fu-berlin.de> wrote:
> > > On Sat, 2023-06-17 at 13:09 +0200, Geert Uytterhoeven wrote:
> > > > On Sat, Jun 17, 2023 at 9:32 AM John Paul Adrian Glaubitz
> > > > <glaubitz@physik.fu-berlin.de> wrote:
> > > > > On Thu, 2023-06-08 at 12:03 +0200, John Paul Adrian Glaubitz wrote:
> > > > > > > > > That is actually safer, as the user can override NR_ONCHIP_DMA_CHANNELS
> > > > > > > > > when configuring his kernel, thus breaking DMA due to an incorrect
> > > > > > > > > value of SH_DMAC_NR_MD_CH.
> > > > > > > > >
> > > > > > > > > Unfortunately we cannot protect against that when using a single DMAC,
> > > > > > > > > as SH_DMAC_NR_MD_CH can be either 4, 6, or 8.
> > > > > > > > >
> > > > > > > > > Perhaps this configuration should be moved from Kconfig to <cpu/dma.h>,
> > > > > > > > > to protect against a user overriding this value?
> > > > > > > >
> > > > > > > > Isn't SH_DMAC_NR_MD_CH already hardwired to the SoC being used?
> > > > > > >
> > > > > > > It depends on CONFIG_NR_ONCHIP_DMA_CHANNELS, while it
> > > > > > > should be fixed based on the SoC.
> > > > > >
> > > > > > I agree. However, I would be fine with merging this patch set first and fixing
> > > > > > this particular issue in a follow-up series.
> > > > >
> > > > > So, my suggestion is to take this series as-is for 6.5, then get the other issues
> > > > > you mentioned fixed for 6.6. I think it's already a gain when these issues are
> > > > > fixed and the kernel boots on the HP Journada 680 again.
> > > >
> > > > Sure, I don't want to block the acceptance of this series at all.
> > > > Thanks!
> > >
> > > Apologies for the late reply. Would you mind adding your Reviewed-by to this patch
> > > before I review and apply the series?
> >
> > With "must be a multiply of two" and "Must be an even number" removed.
> > Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
>
> Thanks. I guess, I will drop the whole
>
> "and clarify that NR_ONCHIP_DMA_CHANNELS must be a multiply of two"
>
> then. Correct?
Correct. Also in the help text.
Thanks!
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
© 2016 - 2026 Red Hat, Inc.