[PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Icenowy Zheng]

The dcache.cva encoding shown in the comments are wrong, it's for
dcache.cval1 (which is restricted to L1) instead.

Fix this in the comment and in the hardcoded instruction.

Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
---
The code is tested on a LiteX SoC with OpenC906 core, and it
successfully boots to Systemd on a SD card connected to LiteSDCard.

This change should be not noticable on C906, but on multi-core C910
cluster it should fixes something. Unfortunately TH1520 seems to be not
so ready to test mainline patches on.

 arch/riscv/include/asm/errata_list.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
index 4180312d2a70..605800bd390e 100644
--- a/arch/riscv/include/asm/errata_list.h
+++ b/arch/riscv/include/asm/errata_list.h
@@ -102,7 +102,7 @@ asm volatile(ALTERNATIVE(						\
  * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
  *   0000001    01001      rs1       000      00000  0001011
  * dcache.cva rs1 (clean, virtual address)
- *   0000001    00100      rs1       000      00000  0001011
+ *   0000001    00101      rs1       000      00000  0001011
  *
  * dcache.cipa rs1 (clean then invalidate, physical address)
  * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
@@ -115,7 +115,7 @@ asm volatile(ALTERNATIVE(						\
  *   0000000    11001     00000      000      00000  0001011
  */
 #define THEAD_inval_A0	".long 0x0265000b"
-#define THEAD_clean_A0	".long 0x0245000b"
+#define THEAD_clean_A0	".long 0x0255000b"
 #define THEAD_flush_A0	".long 0x0275000b"
 #define THEAD_SYNC_S	".long 0x0190000b"
 
-- 
2.38.1


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH 2/2] riscv: errata: prefix T-Head mnemonics with th.

[Icenowy Zheng]

T-Head now maintains some specification for their extended instructions
at [1], in which all instructions are prefixed "th.".

Follow this practice in the kernel comments.

[1] https://github.com/T-head-Semi/thead-extension-spec

Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
---
 arch/riscv/include/asm/errata_list.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
index 605800bd390e..46adc1c9428f 100644
--- a/arch/riscv/include/asm/errata_list.h
+++ b/arch/riscv/include/asm/errata_list.h
@@ -92,25 +92,25 @@ asm volatile(ALTERNATIVE(						\
 #endif
 
 /*
- * dcache.ipa rs1 (invalidate, physical address)
+ * th.dcache.ipa rs1 (invalidate, physical address)
  * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
  *   0000001    01010      rs1       000      00000  0001011
- * dache.iva rs1 (invalida, virtual address)
+ * th.dcache.iva rs1 (invalida, virtual address)
  *   0000001    00110      rs1       000      00000  0001011
  *
- * dcache.cpa rs1 (clean, physical address)
+ * th.dcache.cpa rs1 (clean, physical address)
  * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
  *   0000001    01001      rs1       000      00000  0001011
- * dcache.cva rs1 (clean, virtual address)
+ * th.dcache.cva rs1 (clean, virtual address)
  *   0000001    00101      rs1       000      00000  0001011
  *
- * dcache.cipa rs1 (clean then invalidate, physical address)
+ * th.dcache.cipa rs1 (clean then invalidate, physical address)
  * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
  *   0000001    01011      rs1       000      00000  0001011
- * dcache.civa rs1 (... virtual address)
+ * th.dcache.civa rs1 (... virtual address)
  *   0000001    00111      rs1       000      00000  0001011
  *
- * sync.s (make sure all cache operations finished)
+ * th.sync.s (make sure all cache operations finished)
  * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
  *   0000000    11001     00000      000      00000  0001011
  */
-- 
2.38.1


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Guo Ren]

Good catch. But I hope c906/910 can directly use paddr here. It's
unnecessary to cause software translation & mmu translation here.

diff --git a/arch/riscv/mm/dma-noncoherent.c b/arch/riscv/mm/dma-noncoherent.c
index b0add983530a..30650a0c4481 100644
--- a/arch/riscv/mm/dma-noncoherent.c
+++ b/arch/riscv/mm/dma-noncoherent.c
@@ -24,13 +24,13 @@ void arch_sync_dma_for_device(phys_addr_t paddr,
size_t size,

        switch (dir) {
        case DMA_TO_DEVICE:
-               ALT_CMO_OP(clean, vaddr, size, riscv_cbom_block_size);
+               ALT_CMO_OP(clean, vaddr, paddr, size, riscv_cbom_block_size);
                break;
        case DMA_FROM_DEVICE:
-               ALT_CMO_OP(clean, vaddr, size, riscv_cbom_block_size);
+               ALT_CMO_OP(clean, vaddr, paddr, size, riscv_cbom_block_size);
                break;
        case DMA_BIDIRECTIONAL:
-               ALT_CMO_OP(flush, vaddr, size, riscv_cbom_block_size);
+               ALT_CMO_OP(flush, vaddr, paddr, size, riscv_cbom_block_size);
                break;
        default:
                break;
@@ -47,7 +47,7 @@ void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
                break;
        case DMA_FROM_DEVICE:
        case DMA_BIDIRECTIONAL:
-               ALT_CMO_OP(flush, vaddr, size, riscv_cbom_block_size);
+               ALT_CMO_OP(flush, vaddr, paddr, size, riscv_cbom_block_size);
                break;
        default:
                break;
@@ -57,8 +57,9 @@ void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
 void arch_dma_prep_coherent(struct page *page, size_t size)
 {
        void *flush_addr = page_address(page);
+       phys_addr_t paddr = PFN_PHYS(page_to_pfn(x));

-       ALT_CMO_OP(flush, flush_addr, size, riscv_cbom_block_size);
+       ALT_CMO_OP(flush, flush_addr, paddr, size, riscv_cbom_block_size);
 }

On Tue, Dec 27, 2022 at 10:03 AM Icenowy Zheng <uwu@icenowy.me> wrote:
>
> The dcache.cva encoding shown in the comments are wrong, it's for
> dcache.cval1 (which is restricted to L1) instead.
>
> Fix this in the comment and in the hardcoded instruction.
>
> Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> ---
> The code is tested on a LiteX SoC with OpenC906 core, and it
> successfully boots to Systemd on a SD card connected to LiteSDCard.
>
> This change should be not noticable on C906, but on multi-core C910
> cluster it should fixes something. Unfortunately TH1520 seems to be not
> so ready to test mainline patches on.
>
>  arch/riscv/include/asm/errata_list.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
> index 4180312d2a70..605800bd390e 100644
> --- a/arch/riscv/include/asm/errata_list.h
> +++ b/arch/riscv/include/asm/errata_list.h
> @@ -102,7 +102,7 @@ asm volatile(ALTERNATIVE(                                           \
>   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
>   *   0000001    01001      rs1       000      00000  0001011
>   * dcache.cva rs1 (clean, virtual address)
> - *   0000001    00100      rs1       000      00000  0001011
> + *   0000001    00101      rs1       000      00000  0001011
>   *
>   * dcache.cipa rs1 (clean then invalidate, physical address)
>   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> @@ -115,7 +115,7 @@ asm volatile(ALTERNATIVE(                                           \
>   *   0000000    11001     00000      000      00000  0001011
>   */
>  #define THEAD_inval_A0 ".long 0x0265000b"
> -#define THEAD_clean_A0 ".long 0x0245000b"
> +#define THEAD_clean_A0 ".long 0x0255000b"
>  #define THEAD_flush_A0 ".long 0x0275000b"
>  #define THEAD_SYNC_S   ".long 0x0190000b"
>
> --
> 2.38.1
>


-- 
Best Regards
 Guo Ren

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Guo Ren]

Good catch. But I hope c906/910 can directly use paddr here. It's
unnecessary to cause software translation & mmu translation here.

diff --git a/arch/riscv/mm/dma-noncoherent.c b/arch/riscv/mm/dma-noncoherent.c
index b0add983530a..30650a0c4481 100644
--- a/arch/riscv/mm/dma-noncoherent.c
+++ b/arch/riscv/mm/dma-noncoherent.c
@@ -24,13 +24,13 @@ void arch_sync_dma_for_device(phys_addr_t paddr,
size_t size,

        switch (dir) {
        case DMA_TO_DEVICE:
-               ALT_CMO_OP(clean, vaddr, size, riscv_cbom_block_size);
+               ALT_CMO_OP(clean, vaddr, paddr, size, riscv_cbom_block_size);
                break;
        case DMA_FROM_DEVICE:
-               ALT_CMO_OP(clean, vaddr, size, riscv_cbom_block_size);
+               ALT_CMO_OP(clean, vaddr, paddr, size, riscv_cbom_block_size);
                break;
        case DMA_BIDIRECTIONAL:
-               ALT_CMO_OP(flush, vaddr, size, riscv_cbom_block_size);
+               ALT_CMO_OP(flush, vaddr, paddr, size, riscv_cbom_block_size);
                break;
        default:
                break;
@@ -47,7 +47,7 @@ void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
                break;
        case DMA_FROM_DEVICE:
        case DMA_BIDIRECTIONAL:
-               ALT_CMO_OP(flush, vaddr, size, riscv_cbom_block_size);
+               ALT_CMO_OP(flush, vaddr, paddr, size, riscv_cbom_block_size);
                break;
        default:
                break;
@@ -57,8 +57,9 @@ void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
 void arch_dma_prep_coherent(struct page *page, size_t size)
 {
        void *flush_addr = page_address(page);
+       phys_addr_t paddr = PFN_PHYS(page_to_pfn(x));

-       ALT_CMO_OP(flush, flush_addr, size, riscv_cbom_block_size);
+       ALT_CMO_OP(flush, flush_addr, paddr, size, riscv_cbom_block_size);
 }

On Tue, Dec 27, 2022 at 10:03 AM Icenowy Zheng <uwu@icenowy.me> wrote:
>
> The dcache.cva encoding shown in the comments are wrong, it's for
> dcache.cval1 (which is restricted to L1) instead.
>
> Fix this in the comment and in the hardcoded instruction.
>
> Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> ---
> The code is tested on a LiteX SoC with OpenC906 core, and it
> successfully boots to Systemd on a SD card connected to LiteSDCard.
>
> This change should be not noticable on C906, but on multi-core C910
> cluster it should fixes something. Unfortunately TH1520 seems to be not
> so ready to test mainline patches on.
>
>  arch/riscv/include/asm/errata_list.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
> index 4180312d2a70..605800bd390e 100644
> --- a/arch/riscv/include/asm/errata_list.h
> +++ b/arch/riscv/include/asm/errata_list.h
> @@ -102,7 +102,7 @@ asm volatile(ALTERNATIVE(                                           \
>   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
>   *   0000001    01001      rs1       000      00000  0001011
>   * dcache.cva rs1 (clean, virtual address)
> - *   0000001    00100      rs1       000      00000  0001011
> + *   0000001    00101      rs1       000      00000  0001011
>   *
>   * dcache.cipa rs1 (clean then invalidate, physical address)
>   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> @@ -115,7 +115,7 @@ asm volatile(ALTERNATIVE(                                           \
>   *   0000000    11001     00000      000      00000  0001011
>   */
>  #define THEAD_inval_A0 ".long 0x0265000b"
> -#define THEAD_clean_A0 ".long 0x0245000b"
> +#define THEAD_clean_A0 ".long 0x0255000b"
>  #define THEAD_flush_A0 ".long 0x0275000b"
>  #define THEAD_SYNC_S   ".long 0x0190000b"
>
> --
> 2.38.1
>


-- 
Best Regards
 Guo Ren

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Icenowy Zheng]

在 2022-12-27星期二的 10:47 +0800，Guo Ren写道：
> Good catch. But I hope c906/910 can directly use paddr here. It's
> unnecessary to cause software translation & mmu translation here.

This could be an enhancement independent of this patchset.

In addition, I think this is some remaining of the design of the
original patchset, in which adds Zicbom support and then Xtheadcmo
support as some errata.

> 
> diff --git a/arch/riscv/mm/dma-noncoherent.c b/arch/riscv/mm/dma-
> noncoherent.c
> index b0add983530a..30650a0c4481 100644
> --- a/arch/riscv/mm/dma-noncoherent.c
> +++ b/arch/riscv/mm/dma-noncoherent.c
> @@ -24,13 +24,13 @@ void arch_sync_dma_for_device(phys_addr_t paddr,
> size_t size,
> 
>         switch (dir) {
>         case DMA_TO_DEVICE:
> -               ALT_CMO_OP(clean, vaddr, size,
> riscv_cbom_block_size);
> +               ALT_CMO_OP(clean, vaddr, paddr, size,
> riscv_cbom_block_size);

Maybe the macro should be renamed as ALT_CMO_OP_VPA or similar thing,
that means both VA and PA are provided and the most efficient one will
be chosen, because CMO operations with PA is still some T-Head-only
thing (Zicbom only supports VA).

>                 break;
>         case DMA_FROM_DEVICE:
> -               ALT_CMO_OP(clean, vaddr, size,
> riscv_cbom_block_size);
> +               ALT_CMO_OP(clean, vaddr, paddr, size,
> riscv_cbom_block_size);
>                 break;
>         case DMA_BIDIRECTIONAL:
> -               ALT_CMO_OP(flush, vaddr, size,
> riscv_cbom_block_size);
> +               ALT_CMO_OP(flush, vaddr, paddr, size,
> riscv_cbom_block_size);
>                 break;
>         default:
>                 break;
> @@ -47,7 +47,7 @@ void arch_sync_dma_for_cpu(phys_addr_t paddr,
> size_t size,
>                 break;
>         case DMA_FROM_DEVICE:
>         case DMA_BIDIRECTIONAL:
> -               ALT_CMO_OP(flush, vaddr, size,
> riscv_cbom_block_size);
> +               ALT_CMO_OP(flush, vaddr, paddr, size,
> riscv_cbom_block_size);
>                 break;
>         default:
>                 break;
> @@ -57,8 +57,9 @@ void arch_sync_dma_for_cpu(phys_addr_t paddr,
> size_t size,
>  void arch_dma_prep_coherent(struct page *page, size_t size)
>  {
>         void *flush_addr = page_address(page);
> +       phys_addr_t paddr = PFN_PHYS(page_to_pfn(x));
> 
> -       ALT_CMO_OP(flush, flush_addr, size, riscv_cbom_block_size);
> +       ALT_CMO_OP(flush, flush_addr, paddr, size,
> riscv_cbom_block_size);
>  }
> 
> On Tue, Dec 27, 2022 at 10:03 AM Icenowy Zheng <uwu@icenowy.me>
> wrote:
> > 
> > The dcache.cva encoding shown in the comments are wrong, it's for
> > dcache.cval1 (which is restricted to L1) instead.
> > 
> > Fix this in the comment and in the hardcoded instruction.
> > 
> > Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> > ---
> > The code is tested on a LiteX SoC with OpenC906 core, and it
> > successfully boots to Systemd on a SD card connected to LiteSDCard.
> > 
> > This change should be not noticable on C906, but on multi-core C910
> > cluster it should fixes something. Unfortunately TH1520 seems to be
> > not
> > so ready to test mainline patches on.
> > 
> >  arch/riscv/include/asm/errata_list.h | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/arch/riscv/include/asm/errata_list.h
> > b/arch/riscv/include/asm/errata_list.h
> > index 4180312d2a70..605800bd390e 100644
> > --- a/arch/riscv/include/asm/errata_list.h
> > +++ b/arch/riscv/include/asm/errata_list.h
> > @@ -102,7 +102,7 @@ asm
> > volatile(ALTERNATIVE(                                           \
> >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> >   *   0000001    01001      rs1       000      00000  0001011
> >   * dcache.cva rs1 (clean, virtual address)
> > - *   0000001    00100      rs1       000      00000  0001011
> > + *   0000001    00101      rs1       000      00000  0001011
> >   *
> >   * dcache.cipa rs1 (clean then invalidate, physical address)
> >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> > @@ -115,7 +115,7 @@ asm
> > volatile(ALTERNATIVE(                                           \
> >   *   0000000    11001     00000      000      00000  0001011
> >   */
> >  #define THEAD_inval_A0 ".long 0x0265000b"
> > -#define THEAD_clean_A0 ".long 0x0245000b"
> > +#define THEAD_clean_A0 ".long 0x0255000b"
> >  #define THEAD_flush_A0 ".long 0x0275000b"
> >  #define THEAD_SYNC_S   ".long 0x0190000b"
> > 
> > --
> > 2.38.1
> > 
> 
> 

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Icenowy Zheng]

在 2022-12-27星期二的 10:47 +0800，Guo Ren写道：
> Good catch. But I hope c906/910 can directly use paddr here. It's
> unnecessary to cause software translation & mmu translation here.

This could be an enhancement independent of this patchset.

In addition, I think this is some remaining of the design of the
original patchset, in which adds Zicbom support and then Xtheadcmo
support as some errata.

> 
> diff --git a/arch/riscv/mm/dma-noncoherent.c b/arch/riscv/mm/dma-
> noncoherent.c
> index b0add983530a..30650a0c4481 100644
> --- a/arch/riscv/mm/dma-noncoherent.c
> +++ b/arch/riscv/mm/dma-noncoherent.c
> @@ -24,13 +24,13 @@ void arch_sync_dma_for_device(phys_addr_t paddr,
> size_t size,
> 
>         switch (dir) {
>         case DMA_TO_DEVICE:
> -               ALT_CMO_OP(clean, vaddr, size,
> riscv_cbom_block_size);
> +               ALT_CMO_OP(clean, vaddr, paddr, size,
> riscv_cbom_block_size);

Maybe the macro should be renamed as ALT_CMO_OP_VPA or similar thing,
that means both VA and PA are provided and the most efficient one will
be chosen, because CMO operations with PA is still some T-Head-only
thing (Zicbom only supports VA).

>                 break;
>         case DMA_FROM_DEVICE:
> -               ALT_CMO_OP(clean, vaddr, size,
> riscv_cbom_block_size);
> +               ALT_CMO_OP(clean, vaddr, paddr, size,
> riscv_cbom_block_size);
>                 break;
>         case DMA_BIDIRECTIONAL:
> -               ALT_CMO_OP(flush, vaddr, size,
> riscv_cbom_block_size);
> +               ALT_CMO_OP(flush, vaddr, paddr, size,
> riscv_cbom_block_size);
>                 break;
>         default:
>                 break;
> @@ -47,7 +47,7 @@ void arch_sync_dma_for_cpu(phys_addr_t paddr,
> size_t size,
>                 break;
>         case DMA_FROM_DEVICE:
>         case DMA_BIDIRECTIONAL:
> -               ALT_CMO_OP(flush, vaddr, size,
> riscv_cbom_block_size);
> +               ALT_CMO_OP(flush, vaddr, paddr, size,
> riscv_cbom_block_size);
>                 break;
>         default:
>                 break;
> @@ -57,8 +57,9 @@ void arch_sync_dma_for_cpu(phys_addr_t paddr,
> size_t size,
>  void arch_dma_prep_coherent(struct page *page, size_t size)
>  {
>         void *flush_addr = page_address(page);
> +       phys_addr_t paddr = PFN_PHYS(page_to_pfn(x));
> 
> -       ALT_CMO_OP(flush, flush_addr, size, riscv_cbom_block_size);
> +       ALT_CMO_OP(flush, flush_addr, paddr, size,
> riscv_cbom_block_size);
>  }
> 
> On Tue, Dec 27, 2022 at 10:03 AM Icenowy Zheng <uwu@icenowy.me>
> wrote:
> > 
> > The dcache.cva encoding shown in the comments are wrong, it's for
> > dcache.cval1 (which is restricted to L1) instead.
> > 
> > Fix this in the comment and in the hardcoded instruction.
> > 
> > Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> > ---
> > The code is tested on a LiteX SoC with OpenC906 core, and it
> > successfully boots to Systemd on a SD card connected to LiteSDCard.
> > 
> > This change should be not noticable on C906, but on multi-core C910
> > cluster it should fixes something. Unfortunately TH1520 seems to be
> > not
> > so ready to test mainline patches on.
> > 
> >  arch/riscv/include/asm/errata_list.h | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/arch/riscv/include/asm/errata_list.h
> > b/arch/riscv/include/asm/errata_list.h
> > index 4180312d2a70..605800bd390e 100644
> > --- a/arch/riscv/include/asm/errata_list.h
> > +++ b/arch/riscv/include/asm/errata_list.h
> > @@ -102,7 +102,7 @@ asm
> > volatile(ALTERNATIVE(                                           \
> >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> >   *   0000001    01001      rs1       000      00000  0001011
> >   * dcache.cva rs1 (clean, virtual address)
> > - *   0000001    00100      rs1       000      00000  0001011
> > + *   0000001    00101      rs1       000      00000  0001011
> >   *
> >   * dcache.cipa rs1 (clean then invalidate, physical address)
> >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> > @@ -115,7 +115,7 @@ asm
> > volatile(ALTERNATIVE(                                           \
> >   *   0000000    11001     00000      000      00000  0001011
> >   */
> >  #define THEAD_inval_A0 ".long 0x0265000b"
> > -#define THEAD_clean_A0 ".long 0x0245000b"
> > +#define THEAD_clean_A0 ".long 0x0255000b"
> >  #define THEAD_flush_A0 ".long 0x0275000b"
> >  #define THEAD_SYNC_S   ".long 0x0190000b"
> > 
> > --
> > 2.38.1
> > 
> 
> 

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Heiko Stuebner]

Am Dienstag, 27. Dezember 2022, 03:02:57 CET schrieb Icenowy Zheng:
> The dcache.cva encoding shown in the comments are wrong, it's for
> dcache.cval1 (which is restricted to L1) instead.
> 
> Fix this in the comment and in the hardcoded instruction.
> 
> Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> ---
> The code is tested on a LiteX SoC with OpenC906 core, and it
> successfully boots to Systemd on a SD card connected to LiteSDCard.
> 
> This change should be not noticable on C906, but on multi-core C910
> cluster it should fixes something. Unfortunately TH1520 seems to be not
> so ready to test mainline patches on.
> 
>  arch/riscv/include/asm/errata_list.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
> index 4180312d2a70..605800bd390e 100644
> --- a/arch/riscv/include/asm/errata_list.h
> +++ b/arch/riscv/include/asm/errata_list.h
> @@ -102,7 +102,7 @@ asm volatile(ALTERNATIVE(						\
>   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
>   *   0000001    01001      rs1       000      00000  0001011
>   * dcache.cva rs1 (clean, virtual address)
> - *   0000001    00100      rs1       000      00000  0001011
> + *   0000001    00101      rs1       000      00000  0001011
>   *
>   * dcache.cipa rs1 (clean then invalidate, physical address)
>   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> @@ -115,7 +115,7 @@ asm volatile(ALTERNATIVE(						\
>   *   0000000    11001     00000      000      00000  0001011
>   */
>  #define THEAD_inval_A0	".long 0x0265000b"
> -#define THEAD_clean_A0	".long 0x0245000b"
> +#define THEAD_clean_A0	".long 0x0255000b"

the original encoding came from a chinese document from one of the
t-head repos which only containted the original instruction as
"dcache.cva" [0] ... so I guess some part was lost in translation :-)


It's really great to see that the documentation improved a lot with
that new repo with instruction encodings you mention in patch2.

Using that new repo you mention, the change looks correct
0x4 -> 0x5 for the instruction selection, so

Reviewed-by: Heiko Stuebner <heiko@sntech.de>

Though I'm on xmas vaction right now so can't test it on a board.


Heiko

[0] https://github.com/T-head-Semi/openc906/blob/main/doc/%E7%8E%84%E9%93%81C906%E7%94%A8%E6%88%B7%E6%89%8B%E5%86%8C.pdf
page 233

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Heiko Stuebner]

Am Dienstag, 27. Dezember 2022, 03:02:57 CET schrieb Icenowy Zheng:
> The dcache.cva encoding shown in the comments are wrong, it's for
> dcache.cval1 (which is restricted to L1) instead.
> 
> Fix this in the comment and in the hardcoded instruction.
> 
> Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> ---
> The code is tested on a LiteX SoC with OpenC906 core, and it
> successfully boots to Systemd on a SD card connected to LiteSDCard.
> 
> This change should be not noticable on C906, but on multi-core C910
> cluster it should fixes something. Unfortunately TH1520 seems to be not
> so ready to test mainline patches on.
> 
>  arch/riscv/include/asm/errata_list.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
> index 4180312d2a70..605800bd390e 100644
> --- a/arch/riscv/include/asm/errata_list.h
> +++ b/arch/riscv/include/asm/errata_list.h
> @@ -102,7 +102,7 @@ asm volatile(ALTERNATIVE(						\
>   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
>   *   0000001    01001      rs1       000      00000  0001011
>   * dcache.cva rs1 (clean, virtual address)
> - *   0000001    00100      rs1       000      00000  0001011
> + *   0000001    00101      rs1       000      00000  0001011
>   *
>   * dcache.cipa rs1 (clean then invalidate, physical address)
>   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> @@ -115,7 +115,7 @@ asm volatile(ALTERNATIVE(						\
>   *   0000000    11001     00000      000      00000  0001011
>   */
>  #define THEAD_inval_A0	".long 0x0265000b"
> -#define THEAD_clean_A0	".long 0x0245000b"
> +#define THEAD_clean_A0	".long 0x0255000b"

the original encoding came from a chinese document from one of the
t-head repos which only containted the original instruction as
"dcache.cva" [0] ... so I guess some part was lost in translation :-)


It's really great to see that the documentation improved a lot with
that new repo with instruction encodings you mention in patch2.

Using that new repo you mention, the change looks correct
0x4 -> 0x5 for the instruction selection, so

Reviewed-by: Heiko Stuebner <heiko@sntech.de>

Though I'm on xmas vaction right now so can't test it on a board.


Heiko

[0] https://github.com/T-head-Semi/openc906/blob/main/doc/%E7%8E%84%E9%93%81C906%E7%94%A8%E6%88%B7%E6%89%8B%E5%86%8C.pdf
page 233



_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Conor Dooley]

[-- Attachment #1.1: Type: text/plain, Size: 170 bytes --]

Hi Icenowy,

I can't find this patch on lore or patchwork, (although the replies made
it). Would you please resend it once the discussion has died down?

Thanks,
Conor.


[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

[-- Attachment #2: Type: text/plain, Size: 161 bytes --]

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Conor Dooley]

[-- Attachment #1: Type: text/plain, Size: 170 bytes --]

Hi Icenowy,

I can't find this patch on lore or patchwork, (although the replies made
it). Would you please resend it once the discussion has died down?

Thanks,
Conor.


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Sergey Matyukevich]

> > The dcache.cva encoding shown in the comments are wrong, it's for
> > dcache.cval1 (which is restricted to L1) instead.
> >
> > Fix this in the comment and in the hardcoded instruction.
> >
> > Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> > ---
> > The code is tested on a LiteX SoC with OpenC906 core, and it
> > successfully boots to Systemd on a SD card connected to LiteSDCard.
> >
> > This change should be not noticable on C906, but on multi-core C910
> > cluster it should fixes something. Unfortunately TH1520 seems to be not
> > so ready to test mainline patches on.
> >
> >  arch/riscv/include/asm/errata_list.h | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
> > index 4180312d2a70..605800bd390e 100644
> > --- a/arch/riscv/include/asm/errata_list.h
> > +++ b/arch/riscv/include/asm/errata_list.h
> > @@ -102,7 +102,7 @@ asm volatile(ALTERNATIVE(                                           \
> >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> >   *   0000001    01001      rs1       000      00000  0001011
> >   * dcache.cva rs1 (clean, virtual address)
> > - *   0000001    00100      rs1       000      00000  0001011
> > + *   0000001    00101      rs1       000      00000  0001011
> >   *
> >   * dcache.cipa rs1 (clean then invalidate, physical address)
> >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> > @@ -115,7 +115,7 @@ asm volatile(ALTERNATIVE(                                           \
> >   *   0000000    11001     00000      000      00000  0001011
> >   */
> >  #define THEAD_inval_A0 ".long 0x0265000b"
> > -#define THEAD_clean_A0 ".long 0x0245000b"
> > +#define THEAD_clean_A0 ".long 0x0255000b"
> >  #define THEAD_flush_A0 ".long 0x0275000b"
> >  #define THEAD_SYNC_S   ".long 0x0190000b"

Nice catch ! I experimented with T-Head RVB-ICE board on the up-to-date
upstream kernel, using device tree and some other bits from the vendor
kernel [1]. Without this patch, emmc on this board does not work since
noncoherent dma sync operations use incorrect 'clean' instruction.
Applying this patch fixes emmc operations, so

Tested-by: Sergey Matyukevich <sergey.matyukevich@syntacore.com>

[1] https://github.com/T-head-Semi/linux/tree/linux-5.10.4

Regards,
Sergey

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Sergey Matyukevich]

> > The dcache.cva encoding shown in the comments are wrong, it's for
> > dcache.cval1 (which is restricted to L1) instead.
> >
> > Fix this in the comment and in the hardcoded instruction.
> >
> > Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> > ---
> > The code is tested on a LiteX SoC with OpenC906 core, and it
> > successfully boots to Systemd on a SD card connected to LiteSDCard.
> >
> > This change should be not noticable on C906, but on multi-core C910
> > cluster it should fixes something. Unfortunately TH1520 seems to be not
> > so ready to test mainline patches on.
> >
> >  arch/riscv/include/asm/errata_list.h | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
> > index 4180312d2a70..605800bd390e 100644
> > --- a/arch/riscv/include/asm/errata_list.h
> > +++ b/arch/riscv/include/asm/errata_list.h
> > @@ -102,7 +102,7 @@ asm volatile(ALTERNATIVE(                                           \
> >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> >   *   0000001    01001      rs1       000      00000  0001011
> >   * dcache.cva rs1 (clean, virtual address)
> > - *   0000001    00100      rs1       000      00000  0001011
> > + *   0000001    00101      rs1       000      00000  0001011
> >   *
> >   * dcache.cipa rs1 (clean then invalidate, physical address)
> >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> > @@ -115,7 +115,7 @@ asm volatile(ALTERNATIVE(                                           \
> >   *   0000000    11001     00000      000      00000  0001011
> >   */
> >  #define THEAD_inval_A0 ".long 0x0265000b"
> > -#define THEAD_clean_A0 ".long 0x0245000b"
> > +#define THEAD_clean_A0 ".long 0x0255000b"
> >  #define THEAD_flush_A0 ".long 0x0275000b"
> >  #define THEAD_SYNC_S   ".long 0x0190000b"

Nice catch ! I experimented with T-Head RVB-ICE board on the up-to-date
upstream kernel, using device tree and some other bits from the vendor
kernel [1]. Without this patch, emmc on this board does not work since
noncoherent dma sync operations use incorrect 'clean' instruction.
Applying this patch fixes emmc operations, so

Tested-by: Sergey Matyukevich <sergey.matyukevich@syntacore.com>

[1] https://github.com/T-head-Semi/linux/tree/linux-5.10.4

Regards,
Sergey

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Icenowy Zheng]

在 2022-12-31星期六的 01:12 +0300，Sergey Matyukevich写道：
> > > The dcache.cva encoding shown in the comments are wrong, it's for
> > > dcache.cval1 (which is restricted to L1) instead.
> > > 
> > > Fix this in the comment and in the hardcoded instruction.
> > > 
> > > Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> > > ---
> > > The code is tested on a LiteX SoC with OpenC906 core, and it
> > > successfully boots to Systemd on a SD card connected to
> > > LiteSDCard.
> > > 
> > > This change should be not noticable on C906, but on multi-core
> > > C910
> > > cluster it should fixes something. Unfortunately TH1520 seems to
> > > be not
> > > so ready to test mainline patches on.
> > > 
> > >  arch/riscv/include/asm/errata_list.h | 4 ++--
> > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/arch/riscv/include/asm/errata_list.h
> > > b/arch/riscv/include/asm/errata_list.h
> > > index 4180312d2a70..605800bd390e 100644
> > > --- a/arch/riscv/include/asm/errata_list.h
> > > +++ b/arch/riscv/include/asm/errata_list.h
> > > @@ -102,7 +102,7 @@ asm
> > > volatile(ALTERNATIVE(                                           \
> > >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> > >   *   0000001    01001      rs1       000      00000  0001011
> > >   * dcache.cva rs1 (clean, virtual address)
> > > - *   0000001    00100      rs1       000      00000  0001011
> > > + *   0000001    00101      rs1       000      00000  0001011
> > >   *
> > >   * dcache.cipa rs1 (clean then invalidate, physical address)
> > >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> > > @@ -115,7 +115,7 @@ asm
> > > volatile(ALTERNATIVE(                                           \
> > >   *   0000000    11001     00000      000      00000  0001011
> > >   */
> > >  #define THEAD_inval_A0 ".long 0x0265000b"
> > > -#define THEAD_clean_A0 ".long 0x0245000b"
> > > +#define THEAD_clean_A0 ".long 0x0255000b"
> > >  #define THEAD_flush_A0 ".long 0x0275000b"
> > >  #define THEAD_SYNC_S   ".long 0x0190000b"
> 
> Nice catch ! I experimented with T-Head RVB-ICE board on the up-to-
> date
> upstream kernel, using device tree and some other bits from the
> vendor
> kernel [1]. Without this patch, emmc on this board does not work
> since
> noncoherent dma sync operations use incorrect 'clean' instruction.
> Applying this patch fixes emmc operations, so

Thanks for testing on C910! I don't think this patch will make a
significant difference on C906, so it's valuable!

> 
> Tested-by: Sergey Matyukevich <sergey.matyukevich@syntacore.com>
> 
> [1] https://github.com/T-head-Semi/linux/tree/linux-5.10.4
> 
> Regards,
> Sergey

Re: [PATCH 1/2] riscv: errata: fix T-Head dcache.cva encoding

[Icenowy Zheng]

在 2022-12-31星期六的 01:12 +0300，Sergey Matyukevich写道：
> > > The dcache.cva encoding shown in the comments are wrong, it's for
> > > dcache.cval1 (which is restricted to L1) instead.
> > > 
> > > Fix this in the comment and in the hardcoded instruction.
> > > 
> > > Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> > > ---
> > > The code is tested on a LiteX SoC with OpenC906 core, and it
> > > successfully boots to Systemd on a SD card connected to
> > > LiteSDCard.
> > > 
> > > This change should be not noticable on C906, but on multi-core
> > > C910
> > > cluster it should fixes something. Unfortunately TH1520 seems to
> > > be not
> > > so ready to test mainline patches on.
> > > 
> > >  arch/riscv/include/asm/errata_list.h | 4 ++--
> > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/arch/riscv/include/asm/errata_list.h
> > > b/arch/riscv/include/asm/errata_list.h
> > > index 4180312d2a70..605800bd390e 100644
> > > --- a/arch/riscv/include/asm/errata_list.h
> > > +++ b/arch/riscv/include/asm/errata_list.h
> > > @@ -102,7 +102,7 @@ asm
> > > volatile(ALTERNATIVE(                                           \
> > >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> > >   *   0000001    01001      rs1       000      00000  0001011
> > >   * dcache.cva rs1 (clean, virtual address)
> > > - *   0000001    00100      rs1       000      00000  0001011
> > > + *   0000001    00101      rs1       000      00000  0001011
> > >   *
> > >   * dcache.cipa rs1 (clean then invalidate, physical address)
> > >   * | 31 - 25 | 24 - 20 | 19 - 15 | 14 - 12 | 11 - 7 | 6 - 0 |
> > > @@ -115,7 +115,7 @@ asm
> > > volatile(ALTERNATIVE(                                           \
> > >   *   0000000    11001     00000      000      00000  0001011
> > >   */
> > >  #define THEAD_inval_A0 ".long 0x0265000b"
> > > -#define THEAD_clean_A0 ".long 0x0245000b"
> > > +#define THEAD_clean_A0 ".long 0x0255000b"
> > >  #define THEAD_flush_A0 ".long 0x0275000b"
> > >  #define THEAD_SYNC_S   ".long 0x0190000b"
> 
> Nice catch ! I experimented with T-Head RVB-ICE board on the up-to-
> date
> upstream kernel, using device tree and some other bits from the
> vendor
> kernel [1]. Without this patch, emmc on this board does not work
> since
> noncoherent dma sync operations use incorrect 'clean' instruction.
> Applying this patch fixes emmc operations, so

Thanks for testing on C910! I don't think this patch will make a
significant difference on C906, so it's valuable!

> 
> Tested-by: Sergey Matyukevich <sergey.matyukevich@syntacore.com>
> 
> [1] https://github.com/T-head-Semi/linux/tree/linux-5.10.4
> 
> Regards,
> Sergey


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv