* Re: [PATCH v2 1/4] dt-bindings: clock: renesas,cpg-clocks: Document ZT/ZTR trace clock on R-Mobile A1
From: Marek Vasut @ 2026-04-22 23:33 UTC (permalink / raw)
To: Krzysztof Kozlowski, Marek Vasut
Cc: linux-arm-kernel, Conor Dooley, Geert Uytterhoeven,
Krzysztof Kozlowski, Magnus Damm, Michael Turquette, Rob Herring,
Stephen Boyd, devicetree, linux-clk, linux-kernel,
linux-renesas-soc
In-Reply-To: <20260421-attentive-cockle-of-finesse-e5bfb1@quoll>
On 4/21/26 10:01 AM, Krzysztof Kozlowski wrote:
> On Thu, Apr 16, 2026 at 01:31:38AM +0200, Marek Vasut wrote:
>> Document ZT trace bus and ZTR trace clock on the R-Mobile A1.
>>
>> Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
>> ---
>
> This might affect ABI, because you are basically requiring DTS to have
> additional two clocks. Probably it does not affect anyone, but
> technically is an ABI change, so should be explained in commit msg WHY
> you are doing it and what is the impact, instead of saying WHAT you did.
>
> If there is going to be a new version, please fix it and please keep the
> feedback for future commits.
I hope this is addressed in V3.
^ permalink raw reply
* Re: [PATCH v2 3/4] ARM: dts: renesas: r8a7740: Add ZT/ZTR trace clock on R-Mobile A1
From: Marek Vasut @ 2026-04-22 23:33 UTC (permalink / raw)
To: Krzysztof Kozlowski, Marek Vasut
Cc: linux-arm-kernel, Conor Dooley, Geert Uytterhoeven,
Krzysztof Kozlowski, Magnus Damm, Michael Turquette, Rob Herring,
Stephen Boyd, devicetree, linux-clk, linux-kernel,
linux-renesas-soc
In-Reply-To: <20260421-cherubic-urban-beluga-49ee00@quoll>
On 4/21/26 10:02 AM, Krzysztof Kozlowski wrote:
> On Thu, Apr 16, 2026 at 01:31:40AM +0200, Marek Vasut wrote:
>> Add ZT trace bus and ZTR trace clock on the R-Mobile A1.
>>
>> Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
>> ---
>> Cc: Conor Dooley <conor+dt@kernel.org>
>> Cc: Geert Uytterhoeven <geert+renesas@glider.be>
>> Cc: Krzysztof Kozlowski <krzk+dt@kernel.org>
>> Cc: Magnus Damm <magnus.damm@gmail.com>
>> Cc: Michael Turquette <mturquette@baylibre.com>
>> Cc: Rob Herring <robh@kernel.org>
>> Cc: Stephen Boyd <sboyd@kernel.org>
>> Cc: devicetree@vger.kernel.org
>> Cc: linux-clk@vger.kernel.org
>> Cc: linux-kernel@vger.kernel.org
>> Cc: linux-renesas-soc@vger.kernel.org
>> ---
>> V2: Add ztr/zt clock at the end of the list to match bindings
>> ---
>> arch/arm/boot/dts/renesas/r8a7740.dtsi | 2 +-
>
>> include/dt-bindings/clock/r8a7740-clock.h | 2 ++
>
> This goes to the binding patch.
>
> Didn't you have also a checkpatch warning?
I only got this warning, but the docs 1/4 and includes 3/4 are a
separate patch in this series:
"
WARNING: DT binding docs and includes should be a separate patch. See:
Documentation/devicetree/bindings/submitting-patches.rst
"
^ permalink raw reply
* RE: [PATCH v2 1/3] ASoC: dapm: Fix widget lookup with prefixed names across DAPM contexts
From: Chancel Liu @ 2026-04-23 3:03 UTC (permalink / raw)
To: Frank Li
Cc: lgirdwood@gmail.com, broonie@kernel.org, perex@perex.cz,
tiwai@suse.com, shengjiu.wang@gmail.com, Xiubo.Lee@gmail.com,
festevam@gmail.com, nicoleotsuka@gmail.com,
s.hauer@pengutronix.de, kernel@pengutronix.de,
shumingf@realtek.com, rander.wang@linux.intel.com,
pierre-louis.bossart@linux.dev, linux-sound@vger.kernel.org,
linux-kernel@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
imx@lists.linux.dev, linux-arm-kernel@lists.infradead.org
In-Reply-To: <aeITn0SXl-wqgAyA@lizhi-Precision-Tower-5810>
> subject suggest change to
>
> ASoC: dapm: Use snd_soc_dapm_widget_name_cmp() to fix widget lookup
> failures
>
> Frank
OK, this looks more simplified and clearer.
Regards,
Chancel Liu
> > Currently dapm_find_widget() manually constructs a prefixed widget name
> > based on the provided DAPM context and compares it using strcmp(). This
> > happens to work in most cases because callers usually know which DAPM
> > context the target widget belongs to and pass in the matching DAPM
> > context.
> >
> > However, this assumption breaks when search_other_contexts is enabled.
> > In such cases, callers may intentionally pass a different DAPM context,
> > while searching for a widget that actually belongs to another DAPM
> > context.
> >
> > For example, when searching for a "DAC" widget, the widget belongs to
> > the codec DAPM and be registered with a codec prefix, while the caller
> > passes card->dapm and intends to search across all DAPM contexts. The
> > current implementation incorrectly applies the caller card DAPM causing
> > the lookup to fail even though the widget exists on the card.
> >
> > Use snd_soc_dapm_widget_name_cmp() instead, which compares widget
> names
> > using the widget's own DAPM context and prefix. It fixes widget lookup
> > failures when searching across different DAPM contexts on the card.
> >
> > Fixes: ae4fc532244b ("ASoC: dapm: use component prefix when checking
> widget names")
> > Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
> > Assisted-by: Cody:Claude-3.5-Sonnet
> > ---
> > sound/soc/soc-dapm.c | 13 +------------
> > 1 file changed, 1 insertion(+), 12 deletions(-)
> >
> > diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c
> > index d6192204e613..c5b80d9ed64b 100644
> > --- a/sound/soc/soc-dapm.c
> > +++ b/sound/soc/soc-dapm.c
> > @@ -2906,20 +2906,9 @@ static struct snd_soc_dapm_widget
> *dapm_find_widget(
> > {
> > struct snd_soc_dapm_widget *w;
> > struct snd_soc_dapm_widget *fallback = NULL;
> > - char prefixed_pin[80];
> > - const char *pin_name;
> > - const char *prefix = dapm_prefix(dapm);
> > -
> > - if (prefix) {
> > - snprintf(prefixed_pin, sizeof(prefixed_pin), "%s %s",
> > - prefix, pin);
> > - pin_name = prefixed_pin;
> > - } else {
> > - pin_name = pin;
> > - }
> >
> > for_each_card_widgets(dapm->card, w) {
> > - if (!strcmp(w->name, pin_name)) {
> > + if (!snd_soc_dapm_widget_name_cmp(w, pin)) {
> > if (w->dapm == dapm)
> > return w;
> > else
> > --
> > 2.50.1
> >
^ permalink raw reply
* RE: [PATCH v2 2/3] ASoC: soc-core: Add core support for ignoring suspend on selected DAPM widgets
From: Chancel Liu @ 2026-04-23 3:11 UTC (permalink / raw)
To: Frank Li
Cc: lgirdwood@gmail.com, broonie@kernel.org, perex@perex.cz,
tiwai@suse.com, shengjiu.wang@gmail.com, Xiubo.Lee@gmail.com,
festevam@gmail.com, nicoleotsuka@gmail.com,
s.hauer@pengutronix.de, kernel@pengutronix.de,
shumingf@realtek.com, rander.wang@linux.intel.com,
pierre-louis.bossart@linux.dev, linux-sound@vger.kernel.org,
linux-kernel@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
imx@lists.linux.dev, linux-arm-kernel@lists.infradead.org
In-Reply-To: <aeIWLvMWKv6OmX4h@lizhi-Precision-Tower-5810>
> > Some audio systems require specific DAPM widgets to remain powered
> > during system suspend. Introduce a generic and reusable mechanism in
> > the ASoC core to mark selected DAPM widgets as ignore_suspend.
> >
> > The unified mechanism consists of two parts:
> > 1. Parse and store the name list of widgets to ignore suspend in
> > struct snd_soc_card
> >
> > The list of widgets can be provided either by the machine driver or
> > parsed from Device Tree. Different machines have different routing and
> > power requirements. Each machine can specify its own widgets to ignore
> > suspend through DT property. It enables flexible policy without hard
> > code. A new helper, snd_soc_of_parse_ignore_suspend_widgets() is added
> > for this purpose.
> >
> > 2. Apply ignore_suspend flags during snd_soc_bind_card()
> >
> > After all components have been probed and all DAPM widgets have been
> > registered, snd_soc_bind_card() performs a unified lookup of the
> > configured widget names across all DAPM contexts of the card and marks
> > the matching widgets with ignore_suspend = 1.
> >
> > Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
> > ---
> ...
> > @@ -3294,6 +3298,45 @@ int snd_soc_of_parse_aux_devs(struct
> > snd_soc_card *card, const char *propname) }
> > EXPORT_SYMBOL_GPL(snd_soc_of_parse_aux_devs);
> >
> > +int snd_soc_of_parse_ignore_suspend_widgets(struct snd_soc_card *card,
> > + const char *propname)
>
> propname is fixed value "ignore-suspend-widgets"，needn't this agument.
> you funciton include "ignore_suspend_widgets", suppose only parse property.
> ignore_suspend_widgets
>
> Frank
The motivation for keeping the propname argument is to align with
existing snd_soc_of_parse_*() helper prototypes, which always take
(struct snd_soc_card *card, const char *propname).
I agree that in this case the property name is fixed to
"ignore-suspend-widgets". We can simplify the helper by removing it if
preferred.
Regards,
Chancel Liu
^ permalink raw reply
* [PATCH 1/2] dt-bindings: mfd: rockchip: drop redundant wakeup-source definitions
From: phucduc.bui @ 2026-04-23 4:28 UTC (permalink / raw)
To: Lee Jones, Mark Brown, Liam Girdwood, Rob Herring,
Krzysztof Kozlowski, Conor Dooley, Heiko Stuebner
Cc: Joseph Chen, Chris Zhong, Zhang Qing, David Rau, Animesh Agarwal,
devicetree, linux-sound, linux-rockchip, linux-arm-kernel,
linux-kernel, bui duc phuc
From: bui duc phuc <phucduc.bui@gmail.com>
The 'wakeup-source' property already has its type defined in the core
schema. The Rockchip MFD bindings redundantly specify 'type: boolean'
and provide descriptions which only repeat the generic meaning of the
property. Drop these redundant definitions and descriptions to clean
up the binding files and rely on the core schema instead.
No functional change intended.
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
---
Documentation/devicetree/bindings/mfd/rockchip,rk801.yaml | 5 +----
Documentation/devicetree/bindings/mfd/rockchip,rk805.yaml | 5 +----
Documentation/devicetree/bindings/mfd/rockchip,rk808.yaml | 5 +----
Documentation/devicetree/bindings/mfd/rockchip,rk817.yaml | 5 +----
Documentation/devicetree/bindings/mfd/rockchip,rk818.yaml | 5 +----
5 files changed, 5 insertions(+), 20 deletions(-)
diff --git a/Documentation/devicetree/bindings/mfd/rockchip,rk801.yaml b/Documentation/devicetree/bindings/mfd/rockchip,rk801.yaml
index 7c71447200ba..8e394ab6879b 100644
--- a/Documentation/devicetree/bindings/mfd/rockchip,rk801.yaml
+++ b/Documentation/devicetree/bindings/mfd/rockchip,rk801.yaml
@@ -29,10 +29,7 @@ properties:
description:
Telling whether or not this PMIC is controlling the system power.
- wakeup-source:
- type: boolean
- description:
- Device can be used as a wakeup source.
+ wakeup-source: true
vcc1-supply:
description:
diff --git a/Documentation/devicetree/bindings/mfd/rockchip,rk805.yaml b/Documentation/devicetree/bindings/mfd/rockchip,rk805.yaml
index da2391530c16..548ba1f30712 100644
--- a/Documentation/devicetree/bindings/mfd/rockchip,rk805.yaml
+++ b/Documentation/devicetree/bindings/mfd/rockchip,rk805.yaml
@@ -48,10 +48,7 @@ properties:
system-power-controller: true
- wakeup-source:
- type: boolean
- description:
- Device can be used as a wakeup source.
+ wakeup-source: true
vcc1-supply:
description:
diff --git a/Documentation/devicetree/bindings/mfd/rockchip,rk808.yaml b/Documentation/devicetree/bindings/mfd/rockchip,rk808.yaml
index 50dfffac8fbf..40f8f23afad3 100644
--- a/Documentation/devicetree/bindings/mfd/rockchip,rk808.yaml
+++ b/Documentation/devicetree/bindings/mfd/rockchip,rk808.yaml
@@ -43,10 +43,7 @@ properties:
system-power-controller: true
- wakeup-source:
- type: boolean
- description:
- Device can be used as a wakeup source.
+ wakeup-source: true
vcc1-supply:
description:
diff --git a/Documentation/devicetree/bindings/mfd/rockchip,rk817.yaml b/Documentation/devicetree/bindings/mfd/rockchip,rk817.yaml
index 2cb6d176a84c..b99aaff100fb 100644
--- a/Documentation/devicetree/bindings/mfd/rockchip,rk817.yaml
+++ b/Documentation/devicetree/bindings/mfd/rockchip,rk817.yaml
@@ -55,10 +55,7 @@ properties:
system-power-controller: true
- wakeup-source:
- type: boolean
- description:
- Device can be used as a wakeup source.
+ wakeup-source: true
vcc1-supply:
description:
diff --git a/Documentation/devicetree/bindings/mfd/rockchip,rk818.yaml b/Documentation/devicetree/bindings/mfd/rockchip,rk818.yaml
index 90d944c27ba1..b7af774b1395 100644
--- a/Documentation/devicetree/bindings/mfd/rockchip,rk818.yaml
+++ b/Documentation/devicetree/bindings/mfd/rockchip,rk818.yaml
@@ -43,10 +43,7 @@ properties:
system-power-controller: true
- wakeup-source:
- type: boolean
- description:
- Device can be used as a wakeup source.
+ wakeup-source: true
vcc1-supply:
description:
--
2.43.0
^ permalink raw reply related
* [PATCH 2/2] ASoC: dt-bindings: drop redundant wakeup-source definitions
From: phucduc.bui @ 2026-04-23 4:28 UTC (permalink / raw)
To: Lee Jones, Mark Brown, Liam Girdwood, Rob Herring,
Krzysztof Kozlowski, Conor Dooley, Heiko Stuebner
Cc: Joseph Chen, Chris Zhong, Zhang Qing, David Rau, Animesh Agarwal,
devicetree, linux-sound, linux-rockchip, linux-arm-kernel,
linux-kernel, bui duc phuc
In-Reply-To: <20260423042831.21114-1-phucduc.bui@gmail.com>
From: bui duc phuc <phucduc.bui@gmail.com>
The 'wakeup-source' property already has its type defined in the core
schema. A number of sound binding files redundantly specify the
'type: boolean' and provide descriptions which only repeat the generic
meaning of the property.
Drop these redundant definitions and descriptions to clean up the binding
files and rely on the core schema instead.
No functional change intended.
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
---
Documentation/devicetree/bindings/sound/dialog,da7219.yaml | 5 +----
Documentation/devicetree/bindings/sound/realtek,rt5514.yaml | 4 +---
2 files changed, 2 insertions(+), 7 deletions(-)
diff --git a/Documentation/devicetree/bindings/sound/dialog,da7219.yaml b/Documentation/devicetree/bindings/sound/dialog,da7219.yaml
index 19137abdba3e..fab37cd92e0d 100644
--- a/Documentation/devicetree/bindings/sound/dialog,da7219.yaml
+++ b/Documentation/devicetree/bindings/sound/dialog,da7219.yaml
@@ -45,10 +45,7 @@ properties:
- wakeup
- irq
- wakeup-source:
- type: boolean
- description:
- Flag to indicate this device can wake system (suspend/resume).
+ wakeup-source: true
"#clock-cells":
const: 1
diff --git a/Documentation/devicetree/bindings/sound/realtek,rt5514.yaml b/Documentation/devicetree/bindings/sound/realtek,rt5514.yaml
index 7fbf7739c371..57e5c09c05c0 100644
--- a/Documentation/devicetree/bindings/sound/realtek,rt5514.yaml
+++ b/Documentation/devicetree/bindings/sound/realtek,rt5514.yaml
@@ -48,9 +48,7 @@ properties:
spi-max-frequency: true
- wakeup-source:
- type: boolean
- description: Flag to indicate this device can wake system (suspend/resume).
+ wakeup-source: true
required:
- compatible
--
2.43.0
^ permalink raw reply related
* Re: [PATCH v3 01/11] PCI: Propagate FLR return values to callers
From: Nicolin Chen @ 2026-04-23 4:29 UTC (permalink / raw)
To: Baolu Lu
Cc: Will Deacon, Robin Murphy, Joerg Roedel, Bjorn Helgaas,
Jason Gunthorpe, Rafael J . Wysocki, Len Brown,
Pranjal Shrivastava, Mostafa Saleh, Kevin Tian, linux-arm-kernel,
iommu, linux-kernel, linux-acpi, linux-pci, vsethi, Shuai Xue
In-Reply-To: <cbcbea4e-c857-4b75-a55d-ba3f89cdf22d@linux.intel.com>
On Wed, Apr 22, 2026 at 02:13:49PM +0800, Baolu Lu wrote:
> On 4/17/26 07:28, Nicolin Chen wrote:
> > @@ -4060,7 +4061,7 @@ static int reset_chelsio_generic_dev(struct pci_dev *dev, bool probe)
> > PCI_MSIX_FLAGS_ENABLE |
> > PCI_MSIX_FLAGS_MASKALL);
> > - pcie_flr(dev);
> > + ret = pcie_flr(dev);
>
> It makes more sense to return early here on failure. There is no need to
> perform the subsequent steps if pcie_flr() fails. Would something like
> the following work?
[...]
> @@ -4060,7 +4061,9 @@ static int reset_chelsio_generic_dev(struct pci_dev
> *dev, bool probe)
> PCI_MSIX_FLAGS_ENABLE |
> PCI_MSIX_FLAGS_MASKALL);
>
> - pcie_flr(dev);
> + ret = pcie_flr(dev);
> + if (ret)
> + return ret;
Yea, I was a bit worried to change the behavior, skipping the rest
part. But that's probably the right thing to do.
Thanks
Nicolin
^ permalink raw reply
* Re: [PATCH] arm64: traps: Add a macro to simplify the condition codes check
From: Anshuman Khandual @ 2026-04-23 5:29 UTC (permalink / raw)
To: Jinjie Ruan, catalin.marinas, will, mark.rutland, kees, maz,
ada.coupriediaz, smostafa, leitao, mrigendra.chaubey,
linux-arm-kernel, linux-kernel
In-Reply-To: <20260320082846.1235016-1-ruanjinjie@huawei.com>
On 20/03/26 1:58 PM, Jinjie Ruan wrote:
> Add DEFINE_COND_CHECK macro to define the simple __check_* functions
> to simplify the condition codes check.
>
> No functional changes.
>
> Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
> ---
> arch/arm64/kernel/traps.c | 59 ++++++++++-----------------------------
> 1 file changed, 15 insertions(+), 44 deletions(-)
>
> diff --git a/arch/arm64/kernel/traps.c b/arch/arm64/kernel/traps.c
> index 914282016069..6216fe9e8e42 100644
> --- a/arch/arm64/kernel/traps.c
> +++ b/arch/arm64/kernel/traps.c
> @@ -49,45 +49,21 @@
> #include <asm/system_misc.h>
> #include <asm/sysreg.h>
>
> -static bool __kprobes __check_eq(unsigned long pstate)
> -{
> - return (pstate & PSR_Z_BIT) != 0;
> -}
> -
> -static bool __kprobes __check_ne(unsigned long pstate)
> -{
> - return (pstate & PSR_Z_BIT) == 0;
> -}
> -
> -static bool __kprobes __check_cs(unsigned long pstate)
> -{
> - return (pstate & PSR_C_BIT) != 0;
> -}
> -
> -static bool __kprobes __check_cc(unsigned long pstate)
> -{
> - return (pstate & PSR_C_BIT) == 0;
> -}
> -
> -static bool __kprobes __check_mi(unsigned long pstate)
> -{
> - return (pstate & PSR_N_BIT) != 0;
> -}
> -
> -static bool __kprobes __check_pl(unsigned long pstate)
> -{
> - return (pstate & PSR_N_BIT) == 0;
> -}
> -
> -static bool __kprobes __check_vs(unsigned long pstate)
> -{
> - return (pstate & PSR_V_BIT) != 0;
> -}
> -
> -static bool __kprobes __check_vc(unsigned long pstate)
> -{
> - return (pstate & PSR_V_BIT) == 0;
> -}
> +#define DEFINE_COND_CHECK(name, flag, expected) \
> +static bool __kprobes __check_##name(unsigned long pstate) \
> +{ \
> + return ((pstate & (flag)) != 0) == (expected); \
> +}
> +
> +DEFINE_COND_CHECK(eq, PSR_Z_BIT, true)
> +DEFINE_COND_CHECK(ne, PSR_Z_BIT, false)
> +DEFINE_COND_CHECK(cs, PSR_C_BIT, true)
> +DEFINE_COND_CHECK(cc, PSR_C_BIT, false)
> +DEFINE_COND_CHECK(mi, PSR_N_BIT, true)
> +DEFINE_COND_CHECK(pl, PSR_N_BIT, false)
> +DEFINE_COND_CHECK(vs, PSR_V_BIT, true)
> +DEFINE_COND_CHECK(vc, PSR_V_BIT, false)
> +DEFINE_COND_CHECK(al, 0, false) /* Always true */
(((pstate & 0 == 0) != 0) == false) ---> return true
Although this looks OK but wondering if __check_al() should
be left unchanged for simplicity. OR could all its call sites
be changed assuming an unconditional 'true' return thus later
__check_al() can be dropped.
>
> static bool __kprobes __check_hi(unsigned long pstate)
> {
> @@ -131,11 +107,6 @@ static bool __kprobes __check_le(unsigned long pstate)
> return (temp & PSR_N_BIT) != 0;
> }
>
> -static bool __kprobes __check_al(unsigned long pstate)
> -{
> - return true;
> -}
> -
> /*
> * Note that the ARMv8 ARM calls condition code 0b1111 "nv", but states that
> * it behaves identically to 0b1110 ("al").
^ permalink raw reply
* Re: [REGRESSION] rseq: refactoring in v6.19 broke everyone on arm64 and tcmalloc everywhere
From: Dmitry Vyukov @ 2026-04-23 5:53 UTC (permalink / raw)
To: Jinjie Ruan, linux-man
Cc: Thomas Gleixner, Mark Rutland, Mathias Stearn, Mathieu Desnoyers,
Catalin Marinas, Will Deacon, Boqun Feng, Paul E. McKenney,
Chris Kennelly, regressions, linux-kernel, linux-arm-kernel,
Peter Zijlstra, Ingo Molnar, Blake Oler
In-Reply-To: <c5331cd6-76c8-430d-978e-fcad164e48f6@huawei.com>
On Thu, 23 Apr 2026 at 03:48, Jinjie Ruan <ruanjinjie@huawei.com> wrote:
>
> On 4/23/2026 3:47 AM, Thomas Gleixner wrote:
> > On Wed, Apr 22 2026 at 19:11, Mark Rutland wrote:
> >> On Wed, Apr 22, 2026 at 07:49:30PM +0200, Thomas Gleixner wrote:
> >> Conceptually we just need to use syscall_enter_from_user_mode() and
> >> irqentry_enter_from_user_mode() appropriately.
> >
> > Right. I figured that out.
> >
> >> In practice, I can't use those as-is without introducing the exception
> >> masking problems I just fixed up for irqentry_enter_from_kernel_mode(),
> >> so I'll need to do some similar refactoring first.
> >
> > See below.
> >
> >> I haven't paged everything in yet, so just to cehck, is there anything
> >> that would behave incorrectly if current->rseq.event.user_irq were set
> >> for syscall entry? IIUC it means we'll effectively do the slow path, and
> >> I was wondering if that might be acceptable as a one-line bodge for
> >> stable.
> >
> > It might work, but it's trivial enough to avoid that. See below. That on
> > top of 6.19.y makes the selftests pass too.
>
> This aligns with my thoughts when convert arm64 to generic syscall
> entry. Currently, the arm64 entry code does not distinguish between IRQ
> and syscall entries. It fails to call rseq_note_user_irq_entry() for IRQ
> entries as the generic entry framework does, because arm64 uses
> enter_from_user_mode() exclusively instead of
> irqentry_enter_from_user_mode().
>
> https://lore.kernel.org/all/20260320102620.1336796-10-ruanjinjie@huawei.com/
>
> >
> > Thanks,
> >
> > tglx
> > ---
> > arch/arm64/kernel/entry-common.c | 14 ++++++++++----
> > 1 file changed, 10 insertions(+), 4 deletions(-)
> >
> > --- a/arch/arm64/kernel/entry-common.c
> > +++ b/arch/arm64/kernel/entry-common.c
> > @@ -58,6 +58,12 @@ static void noinstr exit_to_kernel_mode(
> > irqentry_exit(regs, state);
> > }
> >
> > +static __always_inline void arm64_enter_from_user_mode_syscall(struct pt_regs *regs)
> > +{
> > + enter_from_user_mode(regs);
> > + mte_disable_tco_entry(current);
> > +}
> > +
> > /*
> > * Handle IRQ/context state management when entering from user mode.
> > * Before this function is called it is not safe to call regular kernel code,
> > @@ -65,8 +71,8 @@ static void noinstr exit_to_kernel_mode(
> > */
> > static __always_inline void arm64_enter_from_user_mode(struct pt_regs *regs)
> > {
> > - enter_from_user_mode(regs);
> > - mte_disable_tco_entry(current);
> > + arm64_enter_from_user_mode_syscall(regs);
> > + rseq_note_user_irq_entry();
> > }
> >
> > /*
> > @@ -717,7 +723,7 @@ static void noinstr el0_brk64(struct pt_
> >
> > static void noinstr el0_svc(struct pt_regs *regs)
> > {
> > - arm64_enter_from_user_mode(regs);
> > + arm64_enter_from_user_mode_syscall(regs);
> > cortex_a76_erratum_1463225_svc_handler();
> > fpsimd_syscall_enter();
> > local_daif_restore(DAIF_PROCCTX);
> > @@ -869,7 +875,7 @@ static void noinstr el0_cp15(struct pt_r
> >
> > static void noinstr el0_svc_compat(struct pt_regs *regs)
> > {
> > - arm64_enter_from_user_mode(regs);
> > + arm64_enter_from_user_mode_syscall(regs);
> > cortex_a76_erratum_1463225_svc_handler();
> > local_daif_restore(DAIF_PROCCTX);
> > do_el0_svc_compat(regs);
+linux-man
This part of the rseq man page needs to be fixed as well I think. The
kernel no longer reliably provides clearing of rseq_cs on preemption,
right?
https://git.kernel.org/pub/scm/libs/librseq/librseq.git/tree/doc/man/rseq.2#n241
"and set to NULL by the kernel when it restarts an assembly
instruction sequence block,
as well as when the kernel detects that it is preempting or delivering
a signal outside of the range targeted by the rseq_cs."
^ permalink raw reply
* Re: [RFC PATCH v2 1/4] security: ima: call ima_init() again at late_initcall_sync for defered TPM
From: Yeoreum Yun @ 2026-04-23 5:55 UTC (permalink / raw)
To: Mimi Zohar
Cc: linux-security-module, linux-kernel, linux-integrity,
linux-arm-kernel, kvmarm, paul, jmorris, serge, roberto.sassu,
dmitry.kasatkin, eric.snowberg, jarkko, jgg, sudeep.holla, maz,
oupton, joey.gouly, suzuki.poulose, yuzenghui, catalin.marinas,
will, noodles, sebastianene
In-Reply-To: <82803bb3b471898a77084c449b73c7f7b4eb2149.camel@linux.ibm.com>
> On Wed, 2026-04-22 at 20:41 +0100, Yeoreum Yun wrote:
> > > Hi Mimi,
> > >
> > > > On Wed, 2026-04-22 at 17:24 +0100, Yeoreum Yun wrote:
> > > > > To generate the boot_aggregate log in the IMA subsystem with TPM PCR values,
> > > > > the TPM driver must be built as built-in and
> > > > > must be probed before the IMA subsystem is initialized.
> > > > >
> > > > > However, when the TPM device operates over the FF-A protocol using
> > > > > the CRB interface, probing fails and returns -EPROBE_DEFER if
> > > > > the tpm_crb_ffa device — an FF-A device that provides the communication
> > > > > interface to the tpm_crb driver — has not yet been probed.
> > > > >
> > > > > To ensure the TPM device operating over the FF-A protocol with
> > > > > the CRB interface is probed before IMA initialization,
> > > > > the following conditions must be met:
> > > > >
> > > > > 1. The corresponding ffa_device must be registered,
> > > > > which is done via ffa_init().
> > > > >
> > > > > 2. The tpm_crb_driver must successfully probe this device via
> > > > > tpm_crb_ffa_init().
> > > > >
> > > > > 3. The tpm_crb driver using CRB over FF-A can then
> > > > > be probed successfully. (See crb_acpi_add() and
> > > > > tpm_crb_ffa_init() for reference.)
> > > > >
> > > > > Unfortunately, ffa_init(), tpm_crb_ffa_init(), and crb_acpi_driver_init() are
> > > > > all registered with device_initcall, which means crb_acpi_driver_init() may
> > > > > be invoked before ffa_init() and tpm_crb_ffa_init() are completed.
> > > > >
> > > > > When this occurs, probing the TPM device is deferred.
> > > > > However, the deferred probe can happen after the IMA subsystem
> > > > > has already been initialized, since IMA initialization is performed
> > > > > during late_initcall, and deferred_probe_initcall() is performed
> > > > > at the same level.
> > > > >
> > > > > To resolve this, call ima_init() again at late_inicall_sync level
> > > > > so that let IMA not miss TPM PCR value when generating boot_aggregate
> > > > > log though TPM device presents in the system.
> > > > >
> > > > > Signed-off-by: Yeoreum Yun <yeoreum.yun@arm.com>
> > > >
> > > > A lot of change for just detecting whether ima_init() is being called on
> > > > late_initcall or late_initcall_sync(), without any explanation for all the other
> > > > changes (e.g. ima_init_core).
> > > >
> > > > Please just limit the change to just calling ima_init() twice.
> > >
> > > My concern is that ima_update_policy_flags() will be called
> > > when ima_init() is deferred -- not initialised anything.
> > > though functionally, it might be okay however,
> > > I think ima_update_policy_flags() and notifier should work after ima_init()
> > > works logically.
> > >
> > > This change I think not much quite a lot. just wrapper ima_init() with
> > > ima_init_core() with some error handling.
> > >
> > > Am I missing something?
> >
> > Also, if we handle in ima_init() only, but it failed with other reason,
> > we shouldn't call again ima_init() in the late_initcall_sync.
> >
> > To handle this, It wouldn't do in the ima_init() but we need to handle
> > it by caller of ima_init().
>
> Only tpm_default_chip() is being called to set the ima_tpm_chip. On failure,
> instead of going into TPM-bypass mode, return immediately. There are no calls
> to anything else. Just call ima_init() a second time.
I’m not fully convinced this is sufficient.
What I meant is the case where ima_init() fails due to other
initialisation steps, not only tpm_default_chip() (e.g. ima_fs_init()).
If it fails at the late_initcall stage for such reasons, then we
should not call ima_init() again at late_initcall_sync.
For this reason, instead of adding a static variable inside
ima_init(), I think it would be better to manage the state in the
caller and introduce something like an ima_initialised flag. Also, if
initialisation fails for other reasons, the notifier block should be
unregistered.
I’d also like to ask again whether it is fine to call
ima_update_policy_flags() and keep the notifier registered in the
deferred TPM case. While this may be functionally acceptable, it seems
logically questionable to do so when ima_init() has not completed.
There is also a possibility that a deferred case ultimately fails (e.g.
deferred at late_initcall, but then failing at late_initcall_sync
for another reason, even while entering TPM bypass mode). In that case,
it seems more appropriate to handle this state in the caller of
ima_init(), rather than inside ima_init() itself.
Am I still missing something?
--
Sincerely,
Yeoreum Yun
^ permalink raw reply
* [PATCH] crypto: sun8i-ss - avoid hash and rng references
From: Arnd Bergmann @ 2026-04-23 6:55 UTC (permalink / raw)
To: Corentin Labbe, Herbert Xu, David S. Miller, Chen-Yu Tsai,
Jernej Skrabec, Samuel Holland, Eric Biggers
Cc: Arnd Bergmann, Ovidiu Panait, linux-crypto, linux-arm-kernel,
linux-sunxi, linux-kernel
From: Arnd Bergmann <arnd@arndb.de>
While the sun4i-ss and sun8i-ce drivers started selecting CRYPTO_RNG,
the sun8i-ss variant does not, and causes a link failure:
aarch64-linux-ld: drivers/crypto/allwinner/sun8i-ss/sun8i-ss-core.o: in function `sun8i_ss_unregister_algs':
sun8i-ss-core.c:(.text.sun8i_ss_unregister_algs+0x94): undefined reference to `crypto_unregister_rng'
aarch64-linux-ld: drivers/crypto/allwinner/sun8i-ss/sun8i-ss-core.o: in function `sun8i_ss_probe':
sun8i-ss-core.c:(.text.sun8i_ss_probe+0x40c): undefined reference to `crypto_register_rng'
Looking more closely, I see that all of the allwinner crypto drivers have the
same logic where the rng and hash parts of the driver are optional, but then the
generic code is still selected, which is a bit inconsistent, aside from the
missing CRYPTO_RNG select on sun8i-ss.
Change the approach so only the bits that are actually used are built, using
ifdef checks around the optional portions that match the optional references
to the sub-drivers.
Ideally the drivers would get reworked in a way that keeps all the bits
related to the skcipher/ahash/rng codecs in the respective sub-drivers,
rather than having a common driver that knows about all of these.
Fixes: cdadc1435937 ("crypto: cryptomgr - Select algorithm types only when CRYPTO_SELFTESTS")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
The simpler approach would be to add even more 'select' statements
---
drivers/crypto/allwinner/Kconfig | 2 --
drivers/crypto/allwinner/sun4i-ss/sun4i-ss-core.c | 8 ++++++++
drivers/crypto/allwinner/sun8i-ce/sun8i-ce-core.c | 12 ++++++++++++
drivers/crypto/allwinner/sun8i-ss/sun8i-ss-core.c | 12 ++++++++++++
4 files changed, 32 insertions(+), 2 deletions(-)
diff --git a/drivers/crypto/allwinner/Kconfig b/drivers/crypto/allwinner/Kconfig
index 7270e5fbc573..b8e75210a0e3 100644
--- a/drivers/crypto/allwinner/Kconfig
+++ b/drivers/crypto/allwinner/Kconfig
@@ -14,7 +14,6 @@ config CRYPTO_DEV_SUN4I_SS
select CRYPTO_SHA1
select CRYPTO_AES
select CRYPTO_LIB_DES
- select CRYPTO_RNG
select CRYPTO_SKCIPHER
help
Some Allwinner SoC have a crypto accelerator named
@@ -50,7 +49,6 @@ config CRYPTO_DEV_SUN8I_CE
select CRYPTO_CBC
select CRYPTO_AES
select CRYPTO_DES
- select CRYPTO_RNG
depends on CRYPTO_DEV_ALLWINNER
depends on PM
help
diff --git a/drivers/crypto/allwinner/sun4i-ss/sun4i-ss-core.c b/drivers/crypto/allwinner/sun4i-ss/sun4i-ss-core.c
index 58a76e2ba64e..813c4bc6312a 100644
--- a/drivers/crypto/allwinner/sun4i-ss/sun4i-ss-core.c
+++ b/drivers/crypto/allwinner/sun4i-ss/sun4i-ss-core.c
@@ -247,12 +247,14 @@ static int sun4i_ss_debugfs_show(struct seq_file *seq, void *v)
ss_algs[i].stat_req, ss_algs[i].stat_opti, ss_algs[i].stat_fb,
ss_algs[i].stat_bytes);
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN4I_SS_PRNG
case CRYPTO_ALG_TYPE_RNG:
seq_printf(seq, "%s %s reqs=%lu tsize=%lu\n",
ss_algs[i].alg.rng.base.cra_driver_name,
ss_algs[i].alg.rng.base.cra_name,
ss_algs[i].stat_req, ss_algs[i].stat_bytes);
break;
+#endif
case CRYPTO_ALG_TYPE_AHASH:
seq_printf(seq, "%s %s reqs=%lu\n",
ss_algs[i].alg.hash.halg.base.cra_driver_name,
@@ -471,6 +473,7 @@ static int sun4i_ss_probe(struct platform_device *pdev)
goto error_alg;
}
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN4I_SS_PRNG
case CRYPTO_ALG_TYPE_RNG:
err = crypto_register_rng(&ss_algs[i].alg.rng);
if (err) {
@@ -478,6 +481,7 @@ static int sun4i_ss_probe(struct platform_device *pdev)
ss_algs[i].alg.rng.base.cra_name);
}
break;
+#endif
}
}
@@ -497,9 +501,11 @@ static int sun4i_ss_probe(struct platform_device *pdev)
case CRYPTO_ALG_TYPE_AHASH:
crypto_unregister_ahash(&ss_algs[i].alg.hash);
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN4I_SS_PRNG
case CRYPTO_ALG_TYPE_RNG:
crypto_unregister_rng(&ss_algs[i].alg.rng);
break;
+#endif
}
}
error_pm:
@@ -520,9 +526,11 @@ static void sun4i_ss_remove(struct platform_device *pdev)
case CRYPTO_ALG_TYPE_AHASH:
crypto_unregister_ahash(&ss_algs[i].alg.hash);
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN4I_SS_PRNG
case CRYPTO_ALG_TYPE_RNG:
crypto_unregister_rng(&ss_algs[i].alg.rng);
break;
+#endif
}
}
diff --git a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-core.c b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-core.c
index c16bb6ce6ee3..f3b58ed6aed0 100644
--- a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-core.c
+++ b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-core.c
@@ -676,6 +676,7 @@ static int sun8i_ce_debugfs_show(struct seq_file *seq, void *v)
seq_printf(seq, "\tFallback due to SG numbers: %lu\n",
ce_algs[i].stat_fb_maxsg);
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_CE_HASH
case CRYPTO_ALG_TYPE_AHASH:
seq_printf(seq, "%s %s reqs=%lu fallback=%lu\n",
ce_algs[i].alg.hash.base.halg.base.cra_driver_name,
@@ -692,12 +693,15 @@ static int sun8i_ce_debugfs_show(struct seq_file *seq, void *v)
seq_printf(seq, "\tFallback due to SG numbers: %lu\n",
ce_algs[i].stat_fb_maxsg);
break;
+#endif
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_CE_PRNG
case CRYPTO_ALG_TYPE_RNG:
seq_printf(seq, "%s %s reqs=%lu bytes=%lu\n",
ce_algs[i].alg.rng.base.cra_driver_name,
ce_algs[i].alg.rng.base.cra_name,
ce_algs[i].stat_req, ce_algs[i].stat_bytes);
break;
+#endif
}
}
#if defined(CONFIG_CRYPTO_DEV_SUN8I_CE_TRNG) && \
@@ -905,6 +909,7 @@ static int sun8i_ce_register_algs(struct sun8i_ce_dev *ce)
return err;
}
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_CE_HASH
case CRYPTO_ALG_TYPE_AHASH:
id = ce_algs[i].ce_algo_id;
ce_method = ce->variant->alg_hash[id];
@@ -925,6 +930,8 @@ static int sun8i_ce_register_algs(struct sun8i_ce_dev *ce)
return err;
}
break;
+#endif
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_CE_PRNG
case CRYPTO_ALG_TYPE_RNG:
if (ce->variant->prng == CE_ID_NOTSUPP) {
dev_info(ce->dev,
@@ -942,6 +949,7 @@ static int sun8i_ce_register_algs(struct sun8i_ce_dev *ce)
ce_algs[i].ce = NULL;
}
break;
+#endif
default:
ce_algs[i].ce = NULL;
dev_err(ce->dev, "ERROR: tried to register an unknown algo\n");
@@ -963,16 +971,20 @@ static void sun8i_ce_unregister_algs(struct sun8i_ce_dev *ce)
ce_algs[i].alg.skcipher.base.base.cra_name);
crypto_engine_unregister_skcipher(&ce_algs[i].alg.skcipher);
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_CE_HASH
case CRYPTO_ALG_TYPE_AHASH:
dev_info(ce->dev, "Unregister %d %s\n", i,
ce_algs[i].alg.hash.base.halg.base.cra_name);
crypto_engine_unregister_ahash(&ce_algs[i].alg.hash);
break;
+#endif
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_CE_PRNG
case CRYPTO_ALG_TYPE_RNG:
dev_info(ce->dev, "Unregister %d %s\n", i,
ce_algs[i].alg.rng.base.cra_name);
crypto_unregister_rng(&ce_algs[i].alg.rng);
break;
+#endif
}
}
}
diff --git a/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-core.c b/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-core.c
index f45685707e0d..59c9bc45ec0f 100644
--- a/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-core.c
+++ b/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-core.c
@@ -501,12 +501,15 @@ static int sun8i_ss_debugfs_show(struct seq_file *seq, void *v)
seq_printf(seq, "\tFallback due to SG numbers: %lu\n",
ss_algs[i].stat_fb_sgnum);
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_SS_PRNG
case CRYPTO_ALG_TYPE_RNG:
seq_printf(seq, "%s %s reqs=%lu tsize=%lu\n",
ss_algs[i].alg.rng.base.cra_driver_name,
ss_algs[i].alg.rng.base.cra_name,
ss_algs[i].stat_req, ss_algs[i].stat_bytes);
break;
+#endif
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_SS_HASH
case CRYPTO_ALG_TYPE_AHASH:
seq_printf(seq, "%s %s reqs=%lu fallback=%lu\n",
ss_algs[i].alg.hash.base.halg.base.cra_driver_name,
@@ -523,6 +526,7 @@ static int sun8i_ss_debugfs_show(struct seq_file *seq, void *v)
seq_printf(seq, "\tFallback due to SG numbers: %lu\n",
ss_algs[i].stat_fb_sgnum);
break;
+#endif
}
}
return 0;
@@ -707,6 +711,7 @@ static int sun8i_ss_register_algs(struct sun8i_ss_dev *ss)
return err;
}
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_SS_PRNG
case CRYPTO_ALG_TYPE_RNG:
err = crypto_register_rng(&ss_algs[i].alg.rng);
if (err) {
@@ -715,6 +720,8 @@ static int sun8i_ss_register_algs(struct sun8i_ss_dev *ss)
ss_algs[i].ss = NULL;
}
break;
+#endif
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_SS_HASH
case CRYPTO_ALG_TYPE_AHASH:
id = ss_algs[i].ss_algo_id;
ss_method = ss->variant->alg_hash[id];
@@ -735,6 +742,7 @@ static int sun8i_ss_register_algs(struct sun8i_ss_dev *ss)
return err;
}
break;
+#endif
default:
ss_algs[i].ss = NULL;
dev_err(ss->dev, "ERROR: tried to register an unknown algo\n");
@@ -756,16 +764,20 @@ static void sun8i_ss_unregister_algs(struct sun8i_ss_dev *ss)
ss_algs[i].alg.skcipher.base.base.cra_name);
crypto_engine_unregister_skcipher(&ss_algs[i].alg.skcipher);
break;
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_SS_PRNG
case CRYPTO_ALG_TYPE_RNG:
dev_info(ss->dev, "Unregister %d %s\n", i,
ss_algs[i].alg.rng.base.cra_name);
crypto_unregister_rng(&ss_algs[i].alg.rng);
break;
+#endif
+#ifdef CONFIG_CRYPTO_DEV_SUN8I_SS_HASH
case CRYPTO_ALG_TYPE_AHASH:
dev_info(ss->dev, "Unregister %d %s\n", i,
ss_algs[i].alg.hash.base.halg.base.cra_name);
crypto_engine_unregister_ahash(&ss_algs[i].alg.hash);
break;
+#endif
}
}
}
--
2.39.5
^ permalink raw reply related
* Re: [PATCH net v4 0/2] net: airoha: Fix airoha_qdma_cleanup_tx_queue() processing
From: patchwork-bot+netdevbpf @ 2026-04-23 7:20 UTC (permalink / raw)
To: Lorenzo Bianconi
Cc: andrew+netdev, davem, edumazet, kuba, pabeni, horms,
linux-arm-kernel, linux-mediatek, netdev
In-Reply-To: <20260417-airoha_qdma_cleanup_tx_queue-fix-net-v4-0-e04bcc2c9642@kernel.org>
Hello:
This series was applied to netdev/net.git (main)
by Paolo Abeni <pabeni@redhat.com>:
On Fri, 17 Apr 2026 08:36:30 +0200 you wrote:
> Add missing bits in airoha_qdma_cleanup_tx_queue routine.
> Fix airoha_qdma_cleanup_tx_queue processing errors intorduced in commit
> '3f47e67dff1f7 ("net: airoha: Add the capability to consume out-of-order
> DMA tx descriptors")'.
>
> ---
> Changes in v4:
> - Drop patch 2/3 to move entries to queue head in case of DMA mapping
> failure in airoha_dev_xmit().
> - Link to v3: https://lore.kernel.org/r/20260416-airoha_qdma_cleanup_tx_queue-fix-net-v3-0-2b69f5788580@kernel.org
>
> [...]
Here is the summary with links:
- [net,v4,1/2] net: airoha: Move ndesc initialization at end of airoha_qdma_init_tx()
https://git.kernel.org/netdev/net/c/f329924bb494
- [net,v4,2/2] net: airoha: Add missing bits in airoha_qdma_cleanup_tx_queue()
https://git.kernel.org/netdev/net/c/3309965fe44c
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply
* Re: [PATCH v17 1/5] ring-buffer: Flush and stop persistent ring buffer on panic
From: Geert Uytterhoeven @ 2026-04-23 7:28 UTC (permalink / raw)
To: Masami Hiramatsu (Google)
Cc: Steven Rostedt, Catalin Marinas, Will Deacon, Mathieu Desnoyers,
linux-kernel, linux-trace-kernel, Ian Rogers, linux-arm-kernel
In-Reply-To: <177687459412.932171.8121855108122534476.stgit@mhiramat.tok.corp.google.com>
On Wed, 22 Apr 2026 at 18:26, Masami Hiramatsu (Google)
<mhiramat@kernel.org> wrote:
> From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
>
> On real hardware, panic and machine reboot may not flush hardware cache
> to memory. This means the persistent ring buffer, which relies on a
> coherent state of memory, may not have its events written to the buffer
> and they may be lost. Moreover, there may be inconsistency with the
> counters which are used for validation of the integrity of the
> persistent ring buffer which may cause all data to be discarded.
>
> To avoid this issue, stop recording of the ring buffer on panic and
> flush the cache of the ring buffer's memory.
>
> Fixes: e645535a954a ("tracing: Add option to use memmapped memory for trace boot instance")
> Cc: stable@vger.kernel.org
> Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
> Acked-by: Catalin Marinas <catalin.marinas@arm.com>
> arch/m68k/include/asm/Kbuild | 1 +
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> # m68k
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
^ permalink raw reply
* [PATCH v6 1/6] media: dt-bindings: mediatek,vcodec-encoder: Add MT8196
From: Irui Wang @ 2026-04-23 7:33 UTC (permalink / raw)
To: Hans Verkuil, Mauro Carvalho Chehab, Rob Herring,
Matthias Brugger, Krzysztof Kozlowski, angelogioacchino.delregno,
nicolas.dufresne, Tiffany Lin, kyrie wu
Cc: Yunfei Dong, Maoguang Meng, Longfei Wang, Irui Wang,
Project_Global_Chrome_Upstream_Group, linux-media, devicetree,
linux-kernel, linux-arm-kernel, linux-mediatek, Rob Herring (Arm)
In-Reply-To: <20260423073345.27402-1-irui.wang@mediatek.com>
Add support for MT8196 video encoder which uses VCP (Video Co-Processor)
for firmware management. Unlike previous platforms that use SCP/VPU, MT8196
requires VCP to load and execute the video encoding firmware, with the
encoder communicating through VCP to perform encoding operations.
Add the "mediatek,mt8196-vcodec-enc" compatible string and introduce
the "mediatek,vcp" property to reference the VCP device, which is
required for MT8196 encoder operation.
Signed-off-by: Irui Wang <irui.wang@mediatek.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
---
.../media/mediatek,vcodec-encoder.yaml | 22 +++++++++++++++++++
1 file changed, 22 insertions(+)
diff --git a/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml b/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
index ebc615584f92..72698456374a 100644
--- a/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
+++ b/Documentation/devicetree/bindings/media/mediatek,vcodec-encoder.yaml
@@ -24,6 +24,7 @@ properties:
- mediatek,mt8188-vcodec-enc
- mediatek,mt8192-vcodec-enc
- mediatek,mt8195-vcodec-enc
+ - mediatek,mt8196-vcodec-enc
- items:
- const: mediatek,mt8186-vcodec-enc
- const: mediatek,mt8183-vcodec-enc
@@ -58,6 +59,13 @@ properties:
description:
Describes point to scp.
+ mediatek,vcp:
+ $ref: /schemas/types.yaml#/definitions/phandle
+ description:
+ Reference to the VCP (Video Co-Processor) device that loads and executes
+ the video encoding firmware. The encoder communicates with the firmware
+ through VCP to perform encoding operations.
+
power-domains:
maxItems: 1
@@ -76,6 +84,20 @@ required:
- iommus
allOf:
+ - if:
+ properties:
+ compatible:
+ contains:
+ enum:
+ - mediatek,mt8196-vcodec-enc
+
+ then:
+ required:
+ - mediatek,vcp
+ else:
+ properties:
+ mediatek,vcp: false
+
- if:
properties:
compatible:
--
2.45.2
^ permalink raw reply related
* [PATCH v6 0/6] Add support for MT8196 video encoder
From: Irui Wang @ 2026-04-23 7:33 UTC (permalink / raw)
To: Hans Verkuil, Mauro Carvalho Chehab, Rob Herring,
Matthias Brugger, Krzysztof Kozlowski, angelogioacchino.delregno,
nicolas.dufresne, Tiffany Lin, kyrie wu
Cc: Yunfei Dong, Maoguang Meng, Longfei Wang, Irui Wang,
Project_Global_Chrome_Upstream_Group, linux-media, devicetree,
linux-kernel, linux-arm-kernel, linux-mediatek
This patch series add support for MT8196 video encoder.
patch 1: Add dt-bindings.
patch 2~4: Add a new encoder driver interface for new VCP firmware.
patch 5: Add new encoder driver platform data.
patch 6: Add compatible data.
About adding new driver support, the v4l2-compliance report shows:
"Total for mtk-vcodec-enc device /dev/video3: 47, Succeeded: 46, Failed: 1, Warnings: 0"
The 1 Failed case is not caused by current patch set:
fail: v4l2-test-controls.cpp(1171): node->codec_mask & STATEFUL_ENCODER
test VIDIOC_(UN)SUBSCRIBE_EVENT/DQEVENT: FAIL
This patch set depends on "media: mediatek: vcodec: support video decoder in mt8196"[1]
[1] https://patchwork.linuxtv.org/project/linux-media/list/?series=23518
Change in v6:
- Move dt-bindings to patch 1.
- Rebase patch onto decoder's patch set.
- Link to v5: https://patchwork.linuxtv.org/project/linux-media/cover/20260302035244.8994-1-irui.wang@mediatek.com/
Change in v5:
- Rewrite patch5 commit subject.
- Add else statement in patch5.
- Link to v4: https://patchwork.linuxtv.org/project/linux-media/list/?series=21757
Change in v4:
- Rework patch3 commit message.
- Rework patch5 commit with more details.
- Rebase patch onto decoder's patch set.
- Link to v3: https://patchwork.linuxtv.org/project/linux-media/cover/20250814085642.17343-1-kyrie.wu@mediatek.com/
Change in v3:
- Add venc rc buffer alloc failure error handling.
- Add mediatek,vcp property definition in dt-bindning.
Change in v2:
- Add support for VCP encode process.
- Add MT8196 encoder driver platform data.
- Rebase encoder patch onto decoder's patch set.
- Fix some review comments in v1.
Irui Wang (6):
media: dt-bindings: mediatek,vcodec-encoder: Add MT8196
media: mediatek: encoder: Add a new encoder driver interface
media: mediatek: encoder: Add support for common firmware interface
media: mediatek: encoder: Add support for VCP encode process
media: mediatek: encoder: Add a new platform data member
media: mediatek: encoder: Add MT8196 encoder compatible data
.../media/mediatek,vcodec-encoder.yaml | 22 +
.../mediatek/vcodec/common/mtk_vcodec_fw.c | 19 +
.../mediatek/vcodec/common/mtk_vcodec_fw.h | 2 +
.../vcodec/common/mtk_vcodec_fw_priv.h | 1 +
.../vcodec/common/mtk_vcodec_fw_vcp.c | 6 +
.../platform/mediatek/vcodec/encoder/Makefile | 1 +
.../mediatek/vcodec/encoder/mtk_vcodec_enc.c | 36 +-
.../vcodec/encoder/mtk_vcodec_enc_drv.c | 28 +
.../vcodec/encoder/mtk_vcodec_enc_drv.h | 15 +-
.../vcodec/encoder/venc/venc_common_if.c | 685 ++++++++++++++++++
.../vcodec/encoder/venc/venc_h264_if.c | 8 +-
.../mediatek/vcodec/encoder/venc_drv_if.c | 3 +-
.../mediatek/vcodec/encoder/venc_drv_if.h | 11 +-
.../mediatek/vcodec/encoder/venc_ipi_msg.h | 26 +
.../mediatek/vcodec/encoder/venc_vpu_if.c | 47 +-
15 files changed, 858 insertions(+), 52 deletions(-)
create mode 100644 drivers/media/platform/mediatek/vcodec/encoder/venc/venc_common_if.c
--
2.45.2
^ permalink raw reply
* [PATCH v6 4/6] media: mediatek: encoder: Add support for VCP encode process
From: Irui Wang @ 2026-04-23 7:33 UTC (permalink / raw)
To: Hans Verkuil, Mauro Carvalho Chehab, Rob Herring,
Matthias Brugger, Krzysztof Kozlowski, angelogioacchino.delregno,
nicolas.dufresne, Tiffany Lin, kyrie wu
Cc: Yunfei Dong, Maoguang Meng, Longfei Wang, Irui Wang,
Project_Global_Chrome_Upstream_Group, linux-media, devicetree,
linux-kernel, linux-arm-kernel, linux-mediatek
In-Reply-To: <20260423073345.27402-1-irui.wang@mediatek.com>
Adapt the encoder driver to support VCP firmware interface.
Set the encoder driver firmware type to 'VCP'.
Allocate RC buffers using the VCP device.
Send the shared memory address to VCP and map the encoder VSI address
to the CPU address space using the VCP shared memory address.
Signed-off-by: Irui Wang <irui.wang@mediatek.com>
---
.../mediatek/vcodec/common/mtk_vcodec_fw.c | 6 +++++
.../mediatek/vcodec/common/mtk_vcodec_fw.h | 1 +
.../vcodec/common/mtk_vcodec_fw_priv.h | 1 +
.../vcodec/common/mtk_vcodec_fw_vcp.c | 6 +++++
.../vcodec/encoder/mtk_vcodec_enc_drv.c | 3 +++
.../vcodec/encoder/venc/venc_common_if.c | 23 ++++++++++++++-----
.../mediatek/vcodec/encoder/venc_vpu_if.c | 14 ++++++++++-
7 files changed, 47 insertions(+), 7 deletions(-)
diff --git a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.c b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.c
index 0381acceda25..7a504f093bd8 100644
--- a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.c
+++ b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.c
@@ -105,3 +105,9 @@ int mtk_vcodec_fw_get_type(struct mtk_vcodec_fw *fw)
return fw->type;
}
EXPORT_SYMBOL_GPL(mtk_vcodec_fw_get_type);
+
+struct device *mtk_vcodec_fw_get_dev(struct mtk_vcodec_fw *fw)
+{
+ return fw->ops->get_fw_dev(fw);
+}
+EXPORT_SYMBOL_GPL(mtk_vcodec_fw_get_dev);
diff --git a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.h b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.h
index e7304a7dd3e0..56c26b91651e 100644
--- a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.h
+++ b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.h
@@ -43,5 +43,6 @@ int mtk_vcodec_fw_ipi_send(struct mtk_vcodec_fw *fw, int id,
int mtk_vcodec_fw_get_type(struct mtk_vcodec_fw *fw);
int mtk_vcodec_fw_get_ipi(enum mtk_vcodec_fw_type type, int hw_id);
int mtk_vcodec_fw_get_venc_ipi(enum mtk_vcodec_fw_type type);
+struct device *mtk_vcodec_fw_get_dev(struct mtk_vcodec_fw *fw);
#endif /* _MTK_VCODEC_FW_H_ */
diff --git a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_priv.h b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_priv.h
index 0a2a9b010244..710c83c871f4 100644
--- a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_priv.h
+++ b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_priv.h
@@ -29,6 +29,7 @@ struct mtk_vcodec_fw_ops {
int (*ipi_send)(struct mtk_vcodec_fw *fw, int id, void *buf,
unsigned int len, unsigned int wait);
void (*release)(struct mtk_vcodec_fw *fw);
+ struct device *(*get_fw_dev)(struct mtk_vcodec_fw *fw);
};
#if IS_ENABLED(CONFIG_VIDEO_MEDIATEK_VCODEC_VPU)
diff --git a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_vcp.c b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_vcp.c
index 6b69ce44d4bb..2859fe78f67d 100644
--- a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_vcp.c
+++ b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_vcp.c
@@ -500,6 +500,11 @@ static void mtk_vcodec_vcp_release(struct mtk_vcodec_fw *fw)
}
+static struct device *mtk_vcodec_vcp_get_fw_dev(struct mtk_vcodec_fw *fw)
+{
+ return fw->vcp->vcp_device->dev;
+}
+
static const struct mtk_vcodec_fw_ops mtk_vcodec_vcp_msg = {
.load_firmware = mtk_vcodec_vcp_load_firmware,
.get_vdec_capa = mtk_vcodec_vcp_get_vdec_capa,
@@ -508,6 +513,7 @@ static const struct mtk_vcodec_fw_ops mtk_vcodec_vcp_msg = {
.ipi_register = mtk_vcodec_vcp_set_ipi_register,
.ipi_send = mtk_vcodec_vcp_ipi_send,
.release = mtk_vcodec_vcp_release,
+ .get_fw_dev = mtk_vcodec_vcp_get_fw_dev,
};
struct mtk_vcodec_fw *mtk_vcodec_fw_vcp_init(void *priv, enum mtk_vcodec_fw_use fw_use)
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c
index 4e4541b2fc8e..2f6ee0cd15e3 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c
@@ -262,6 +262,9 @@ static int mtk_vcodec_probe(struct platform_device *pdev)
} else if (!of_property_read_u32(pdev->dev.of_node, "mediatek,scp",
&rproc_phandle)) {
fw_type = SCP;
+ } else if (!of_property_read_u32(pdev->dev.of_node, "mediatek,vcp",
+ &rproc_phandle)) {
+ fw_type = VCP;
} else {
dev_err(&pdev->dev, "[MTK VCODEC] Could not get venc IPI device");
return -ENODEV;
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_common_if.c b/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_common_if.c
index 050b827f0fd0..d981155aeb8c 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_common_if.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_common_if.c
@@ -480,8 +480,13 @@ static void venc_free_rc_buf(struct venc_inst *inst,
{
int i;
struct device *dev;
+ struct mtk_vcodec_fw *fw = inst->ctx->dev->fw_handler;
+
+ if (mtk_vcodec_fw_get_type(fw) == VCP)
+ dev = mtk_vcodec_fw_get_dev(fw);
+ else
+ dev = &inst->ctx->dev->plat_dev->dev;
- dev = &inst->ctx->dev->plat_dev->dev;
mtk_venc_mem_free(inst, dev, &bufs->rc_code);
for (i = 0; i < core_num; i++)
@@ -530,12 +535,18 @@ static int venc_alloc_rc_buf(struct venc_inst *inst,
struct device *dev;
void *tmp_va;
- dev = &inst->ctx->dev->plat_dev->dev;
- if (mtk_venc_mem_alloc(inst, dev, &bufs->rc_code))
- return -ENOMEM;
+ if (mtk_vcodec_fw_get_type(fw) == VCP) {
+ dev = mtk_vcodec_fw_get_dev(fw);
+ if (mtk_venc_mem_alloc(inst, dev, &bufs->rc_code))
+ return -ENOMEM;
+ } else {
+ dev = &inst->ctx->dev->plat_dev->dev;
+ if (mtk_venc_mem_alloc(inst, dev, &bufs->rc_code))
+ return -ENOMEM;
- tmp_va = mtk_vcodec_fw_map_dm_addr(fw, bufs->rc_code.pa);
- memcpy(bufs->rc_code.va, tmp_va, bufs->rc_code.size);
+ tmp_va = mtk_vcodec_fw_map_dm_addr(fw, bufs->rc_code.pa);
+ memcpy(bufs->rc_code.va, tmp_va, bufs->rc_code.size);
+ }
for (i = 0; i < core_num; i++) {
if (mtk_venc_mem_alloc(inst, dev, &bufs->rc_info[i]))
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/venc_vpu_if.c b/drivers/media/platform/mediatek/vcodec/encoder/venc_vpu_if.c
index 7772b8442ebc..0f4693e04a9f 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/venc_vpu_if.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/venc_vpu_if.c
@@ -8,13 +8,23 @@
#include "venc_ipi_msg.h"
#include "venc_vpu_if.h"
+#define VSI_OFFSET_MASK 0x0FFFFFFF
+
static void handle_enc_init_msg(struct venc_vpu_inst *vpu, const void *data)
{
const struct venc_vpu_ipi_msg_init_comm *msg = data;
struct mtk_vcodec_fw *fw = vpu->ctx->dev->fw_handler;
+ u64 pa_start, vsi_offset;
vpu->inst_addr = msg->init_ack.vpu_inst_addr;
- vpu->vsi = mtk_vcodec_fw_map_dm_addr(fw, vpu->inst_addr);
+
+ if (mtk_vcodec_fw_get_type(fw) == VCP) {
+ pa_start = (u64)fw->vcp->iova_addr;
+ vsi_offset = (msg->vpu_vsi_addr & VSI_OFFSET_MASK) - (pa_start & VSI_OFFSET_MASK);
+ vpu->vsi = mtk_vcodec_fw_map_dm_addr(fw, ENCODER_MEM) + vsi_offset;
+ } else {
+ vpu->vsi = mtk_vcodec_fw_map_dm_addr(fw, msg->vpu_vsi_addr);
+ }
/* Firmware version field value is unspecified on MT8173. */
if (mtk_vcodec_fw_get_type(fw) == VPU)
@@ -155,6 +165,8 @@ int vpu_enc_init(struct venc_vpu_inst *vpu)
out.base.venc_inst = (unsigned long)vpu;
if (MTK_ENC_DRV_IS_COMM(vpu->ctx)) {
out.codec_type = vpu->ctx->q_data[MTK_Q_DATA_DST].fmt->fourcc;
+ if (mtk_vcodec_fw_get_type(vpu->ctx->dev->fw_handler) == VCP)
+ out.shared_iova = vpu->ctx->dev->fw_handler->vcp->iova_addr;
msg_size = sizeof(struct venc_ap_ipi_msg_init_comm);
} else {
msg_size = sizeof(struct venc_ap_ipi_msg_init);
--
2.45.2
^ permalink raw reply related
* [PATCH v6 5/6] media: mediatek: encoder: Add a new platform data member
From: Irui Wang @ 2026-04-23 7:33 UTC (permalink / raw)
To: Hans Verkuil, Mauro Carvalho Chehab, Rob Herring,
Matthias Brugger, Krzysztof Kozlowski, angelogioacchino.delregno,
nicolas.dufresne, Tiffany Lin, kyrie wu
Cc: Yunfei Dong, Maoguang Meng, Longfei Wang, Irui Wang,
Project_Global_Chrome_Upstream_Group, linux-media, devicetree,
linux-kernel, linux-arm-kernel, linux-mediatek
In-Reply-To: <20260423073345.27402-1-irui.wang@mediatek.com>
Add a new platform data member to indicate each encoder IC, so that the
get chip name function by 'of_device_is_compatible' can be removed.
Signed-off-by: Irui Wang <irui.wang@mediatek.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
---
.../mediatek/vcodec/encoder/mtk_vcodec_enc.c | 22 ++-----------------
.../vcodec/encoder/mtk_vcodec_enc_drv.c | 6 +++++
.../vcodec/encoder/mtk_vcodec_enc_drv.h | 2 ++
3 files changed, 10 insertions(+), 20 deletions(-)
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc.c b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc.c
index f0344888f2cf..b2f911746c01 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc.c
@@ -198,33 +198,15 @@ static int vidioc_enum_fmt_vid_out(struct file *file, void *priv,
pdata->num_output_formats);
}
-static int mtk_vcodec_enc_get_chip_name(struct mtk_vcodec_enc_ctx *ctx)
-{
- struct device *dev = &ctx->dev->plat_dev->dev;
-
- if (of_device_is_compatible(dev->of_node, "mediatek,mt8173-vcodec-enc"))
- return 8173;
- else if (of_device_is_compatible(dev->of_node, "mediatek,mt8183-vcodec-enc"))
- return 8183;
- else if (of_device_is_compatible(dev->of_node, "mediatek,mt8192-vcodec-enc"))
- return 8192;
- else if (of_device_is_compatible(dev->of_node, "mediatek,mt8195-vcodec-enc"))
- return 8195;
- else if (of_device_is_compatible(dev->of_node, "mediatek,mt8188-vcodec-enc"))
- return 8188;
- else
- return 8173;
-}
-
static int vidioc_venc_querycap(struct file *file, void *priv,
struct v4l2_capability *cap)
{
struct mtk_vcodec_enc_ctx *ctx = file_to_enc_ctx(file);
+ const struct mtk_vcodec_enc_pdata *pdata = ctx->dev->venc_pdata;
struct device *dev = &ctx->dev->plat_dev->dev;
- int platform_name = mtk_vcodec_enc_get_chip_name(ctx);
strscpy(cap->driver, dev->driver->name, sizeof(cap->driver));
- snprintf(cap->card, sizeof(cap->card), "MT%d video encoder", platform_name);
+ snprintf(cap->card, sizeof(cap->card), "MT%d video encoder", pdata->venc_model_num);
return 0;
}
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c
index 2f6ee0cd15e3..7b644f55963b 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c
@@ -390,6 +390,7 @@ static int mtk_vcodec_probe(struct platform_device *pdev)
}
static const struct mtk_vcodec_enc_pdata mt8173_avc_pdata = {
+ .venc_model_num = 8173,
.capture_formats = mtk_video_formats_capture_h264,
.num_capture_formats = ARRAY_SIZE(mtk_video_formats_capture_h264),
.output_formats = mtk_video_formats_output,
@@ -400,6 +401,7 @@ static const struct mtk_vcodec_enc_pdata mt8173_avc_pdata = {
};
static const struct mtk_vcodec_enc_pdata mt8173_vp8_pdata = {
+ .venc_model_num = 8173,
.capture_formats = mtk_video_formats_capture_vp8,
.num_capture_formats = ARRAY_SIZE(mtk_video_formats_capture_vp8),
.output_formats = mtk_video_formats_output,
@@ -410,6 +412,7 @@ static const struct mtk_vcodec_enc_pdata mt8173_vp8_pdata = {
};
static const struct mtk_vcodec_enc_pdata mt8183_pdata = {
+ .venc_model_num = 8183,
.uses_ext = true,
.capture_formats = mtk_video_formats_capture_h264,
.num_capture_formats = ARRAY_SIZE(mtk_video_formats_capture_h264),
@@ -421,6 +424,7 @@ static const struct mtk_vcodec_enc_pdata mt8183_pdata = {
};
static const struct mtk_vcodec_enc_pdata mt8188_pdata = {
+ .venc_model_num = 8188,
.uses_ext = true,
.capture_formats = mtk_video_formats_capture_h264,
.num_capture_formats = ARRAY_SIZE(mtk_video_formats_capture_h264),
@@ -433,6 +437,7 @@ static const struct mtk_vcodec_enc_pdata mt8188_pdata = {
};
static const struct mtk_vcodec_enc_pdata mt8192_pdata = {
+ .venc_model_num = 8192,
.uses_ext = true,
.capture_formats = mtk_video_formats_capture_h264,
.num_capture_formats = ARRAY_SIZE(mtk_video_formats_capture_h264),
@@ -444,6 +449,7 @@ static const struct mtk_vcodec_enc_pdata mt8192_pdata = {
};
static const struct mtk_vcodec_enc_pdata mt8195_pdata = {
+ .venc_model_num = 8195,
.uses_ext = true,
.capture_formats = mtk_video_formats_capture_h264,
.num_capture_formats = ARRAY_SIZE(mtk_video_formats_capture_h264),
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
index 0529564027c4..769fb5009964 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
@@ -21,6 +21,7 @@
/**
* struct mtk_vcodec_enc_pdata - compatible data for each IC
*
+ * @venc_model_num: encoder model number
* @uses_ext: whether the encoder uses the extended firmware messaging format
* @min_bitrate: minimum supported encoding bitrate
* @max_bitrate: maximum supported encoding bitrate
@@ -33,6 +34,7 @@
* @uses_common_fw_iface: whether the encoder uses common driver interface
*/
struct mtk_vcodec_enc_pdata {
+ u16 venc_model_num;
bool uses_ext;
u64 min_bitrate;
u64 max_bitrate;
--
2.45.2
^ permalink raw reply related
* [PATCH v6 2/6] media: mediatek: encoder: Add a new encoder driver interface
From: Irui Wang @ 2026-04-23 7:33 UTC (permalink / raw)
To: Hans Verkuil, Mauro Carvalho Chehab, Rob Herring,
Matthias Brugger, Krzysztof Kozlowski, angelogioacchino.delregno,
nicolas.dufresne, Tiffany Lin, kyrie wu
Cc: Yunfei Dong, Maoguang Meng, Longfei Wang, Irui Wang,
Project_Global_Chrome_Upstream_Group, linux-media, devicetree,
linux-kernel, linux-arm-kernel, linux-mediatek
In-Reply-To: <20260423073345.27402-1-irui.wang@mediatek.com>
Introduce a new encoder kernel driver interface to ensure compatibility
with the updated encoder software driver running in firmware.
The new driver interface is expected to support more encoder formats,
share more encode parameters between kernel and firmware.
Signed-off-by: Irui Wang <irui.wang@mediatek.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
---
.../mediatek/vcodec/common/mtk_vcodec_fw.c | 13 +
.../mediatek/vcodec/common/mtk_vcodec_fw.h | 1 +
.../platform/mediatek/vcodec/encoder/Makefile | 1 +
.../mediatek/vcodec/encoder/mtk_vcodec_enc.c | 14 +-
.../vcodec/encoder/mtk_vcodec_enc_drv.h | 8 +-
.../vcodec/encoder/venc/venc_common_if.c | 674 ++++++++++++++++++
.../vcodec/encoder/venc/venc_h264_if.c | 8 +-
.../mediatek/vcodec/encoder/venc_drv_if.h | 11 +-
8 files changed, 712 insertions(+), 18 deletions(-)
create mode 100644 drivers/media/platform/mediatek/vcodec/encoder/venc/venc_common_if.c
diff --git a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.c b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.c
index 4ed7639dfa30..0381acceda25 100644
--- a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.c
+++ b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.c
@@ -18,6 +18,19 @@ int mtk_vcodec_fw_get_ipi(enum mtk_vcodec_fw_type type, int hw_id)
}
EXPORT_SYMBOL_GPL(mtk_vcodec_fw_get_ipi);
+int mtk_vcodec_fw_get_venc_ipi(enum mtk_vcodec_fw_type type)
+{
+ switch (type) {
+ case SCP:
+ return SCP_IPI_VENC_H264;
+ case VCP:
+ return VCP_IPI_ENCODER;
+ default:
+ return -EINVAL;
+ }
+}
+EXPORT_SYMBOL_GPL(mtk_vcodec_fw_get_venc_ipi);
+
struct mtk_vcodec_fw *mtk_vcodec_fw_select(void *priv, enum mtk_vcodec_fw_type type,
enum mtk_vcodec_fw_use fw_use)
{
diff --git a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.h b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.h
index 142e2e87905c..e7304a7dd3e0 100644
--- a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.h
+++ b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw.h
@@ -42,5 +42,6 @@ int mtk_vcodec_fw_ipi_send(struct mtk_vcodec_fw *fw, int id,
void *buf, unsigned int len, unsigned int wait);
int mtk_vcodec_fw_get_type(struct mtk_vcodec_fw *fw);
int mtk_vcodec_fw_get_ipi(enum mtk_vcodec_fw_type type, int hw_id);
+int mtk_vcodec_fw_get_venc_ipi(enum mtk_vcodec_fw_type type);
#endif /* _MTK_VCODEC_FW_H_ */
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/Makefile b/drivers/media/platform/mediatek/vcodec/encoder/Makefile
index e621b5b7e5e6..9d3229d56e39 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/Makefile
+++ b/drivers/media/platform/mediatek/vcodec/encoder/Makefile
@@ -4,6 +4,7 @@ obj-$(CONFIG_VIDEO_MEDIATEK_VCODEC) += mtk-vcodec-enc.o
mtk-vcodec-enc-y := venc/venc_vp8_if.o \
venc/venc_h264_if.o \
+ venc/venc_common_if.o \
mtk_vcodec_enc.o \
mtk_vcodec_enc_drv.o \
mtk_vcodec_enc_pm.o \
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc.c b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc.c
index 48cb5dded70a..f0344888f2cf 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc.c
@@ -81,11 +81,11 @@ static int vidioc_venc_s_ctrl(struct v4l2_ctrl *ctrl)
break;
case V4L2_CID_MPEG_VIDEO_H264_PROFILE:
mtk_v4l2_venc_dbg(2, ctx, "V4L2_CID_MPEG_VIDEO_H264_PROFILE val = %d", ctrl->val);
- p->h264_profile = ctrl->val;
+ p->profile = ctrl->val;
break;
case V4L2_CID_MPEG_VIDEO_H264_LEVEL:
mtk_v4l2_venc_dbg(2, ctx, "V4L2_CID_MPEG_VIDEO_H264_LEVEL val = %d", ctrl->val);
- p->h264_level = ctrl->val;
+ p->level = ctrl->val;
break;
case V4L2_CID_MPEG_VIDEO_H264_I_PERIOD:
mtk_v4l2_venc_dbg(2, ctx, "V4L2_CID_MPEG_VIDEO_H264_I_PERIOD val = %d", ctrl->val);
@@ -385,8 +385,8 @@ static void mtk_venc_set_param(struct mtk_vcodec_enc_ctx *ctx,
mtk_v4l2_venc_err(ctx, "Unsupported fourcc =%d", q_data_src->fmt->fourcc);
break;
}
- param->h264_profile = enc_params->h264_profile;
- param->h264_level = enc_params->h264_level;
+ param->profile = enc_params->profile;
+ param->level = enc_params->level;
/* Config visible resolution */
param->width = q_data_src->visible_width;
@@ -402,8 +402,8 @@ static void mtk_venc_set_param(struct mtk_vcodec_enc_ctx *ctx,
mtk_v4l2_venc_dbg(0, ctx,
"fmt 0x%x, P/L %d/%d w/h %d/%d buf %d/%d fps/bps %d/%d gop %d i_per %d",
- param->input_yuv_fmt, param->h264_profile,
- param->h264_level, param->width, param->height,
+ param->input_yuv_fmt, param->profile,
+ param->level, param->width, param->height,
param->buf_width, param->buf_height,
param->frm_rate, param->bitrate,
param->gop_size, param->intra_period);
@@ -1157,6 +1157,8 @@ static void mtk_venc_worker(struct work_struct *work)
frm_buf.fb_addr[i].size =
(size_t)src_buf->vb2_buf.planes[i].length;
}
+ frm_buf.num_planes = src_buf->vb2_buf.num_planes;
+
bs_buf.va = vb2_plane_vaddr(&dst_buf->vb2_buf, 0);
bs_buf.dma_addr = vb2_dma_contig_plane_dma_addr(&dst_buf->vb2_buf, 0);
bs_buf.size = (size_t)dst_buf->vb2_buf.planes[0].length;
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
index 0cddfa13594f..53369adc083b 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
@@ -69,8 +69,8 @@ enum mtk_encode_param {
* @framerate_denom: frame rate denominator. ex: framerate_num=30 and
* framerate_denom=1 means FPS is 30
* @h264_max_qp: Max value for H.264 quantization parameter
- * @h264_profile: V4L2 defined H.264 profile
- * @h264_level: V4L2 defined H.264 level
+ * @profile: V4L2 defined profile
+ * @level: V4L2 defined level
* @force_intra: force/insert intra frame
*/
struct mtk_enc_params {
@@ -84,8 +84,8 @@ struct mtk_enc_params {
unsigned int framerate_num;
unsigned int framerate_denom;
unsigned int h264_max_qp;
- unsigned int h264_profile;
- unsigned int h264_level;
+ unsigned int profile;
+ unsigned int level;
unsigned int force_intra;
};
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_common_if.c b/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_common_if.c
new file mode 100644
index 000000000000..050b827f0fd0
--- /dev/null
+++ b/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_common_if.c
@@ -0,0 +1,674 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2025 MediaTek Inc.
+ */
+
+#include "../mtk_vcodec_enc.h"
+#include "../mtk_vcodec_enc_drv.h"
+#include "../venc_drv_base.h"
+#include "../venc_drv_if.h"
+#include "../venc_vpu_if.h"
+#include "../../common/mtk_vcodec_intr.h"
+#include "../../common/mtk_vcodec_util.h"
+
+#define SEQ_HEADER_SIZE 1024
+#define PPS_SIZE 128
+#define MAX_DPB_SIZE 16
+#define MAX_VENC_CORE 3
+#define VENC_CONFIG_LENGTH 115
+#define VENC_CONFIG_DATA 128
+#define VENC_PIC_BITSTREAM_BYTE_CNT 0x0098
+
+/**
+ * enum venc_bs_mode - encode bitstream mode
+ * @VENC_BS_MODE_SPS: encode sps
+ * @VENC_BS_MODE_PPS: encode pps
+ * @VENC_BS_MODE_VPS: encode vps
+ * @VENC_BS_MODE_SEQ_HDR: encode sequence header
+ * @VENC_BS_MODE_FRAME: encode frame
+ * @VENC_BS_MODE_FRAME_FINAL: encode final frame
+ * @VENC_BS_MODE_MAX: max value
+ */
+enum venc_bs_mode {
+ VENC_BS_MODE_SPS = 0,
+ VENC_BS_MODE_PPS,
+ VENC_BS_MODE_VPS,
+ VENC_BS_MODE_SEQ_HDR,
+ VENC_BS_MODE_FRAME,
+ VENC_BS_MODE_FRAME_FINAL,
+ VENC_BS_MODE_MAX
+};
+
+/**
+ * struct venc_config - Structure for encoder configuration
+ * AP-W/R : AP is writer/reader on this item
+ * MCU-W/R: MCU is write/reader on this item
+ * @input_fourcc: input format fourcc
+ * @bitrate: target bitrate (in bps)
+ * @pic_w: visible width of resolution
+ * @pic_h: visible height of resolution
+ * @buf_w: buffer alignment width of resolution
+ * @buf_h: buffer alignment height of resolution
+ * @gop_size: group of picture size (IDR frame period)
+ * @intra_period: I frame period
+ * @framerate: frame rate in fps
+ * @profile: profile_idc in SPS
+ * @level: level_idc in SPS
+ * @core_num: encoder core num
+ * @dpb_size: encode dpb size
+ * @reserved: reserved fields config
+ */
+struct venc_config {
+ __u32 input_fourcc;
+ __u32 bitrate;
+ __u32 pic_w;
+ __u32 pic_h;
+ __u32 buf_w;
+ __u32 buf_h;
+ __u32 gop_size;
+ __u32 intra_period;
+ __u32 framerate;
+ __u32 profile;
+ __u32 level;
+ __u32 core_num;
+ __u32 dpb_size;
+ __u32 reserved[VENC_CONFIG_LENGTH];
+};
+
+/**
+ * struct venc_config_data - Structure for configuration data
+ * @config_data: extended configuration data besides the basic configuration
+ */
+struct venc_config_data {
+ unsigned int config_data[VENC_CONFIG_DATA];
+};
+
+/**
+ * struct venc_work_buf - Structure for working buffer information
+ * AP-W/R : AP is writer/reader on this item
+ * MCU-W/R: MCU is write/reader on this item
+ * @iova: IO virtual address
+ * @pa: physical address
+ * @pa_64: for 64bit pa padding
+ * @va: virtual address
+ * @va_padding: for 64bit va padding
+ * @size: buffer size
+ * @size_padding: for 64bit size padding
+ */
+struct venc_work_buf {
+ unsigned long long iova;
+ union {
+ unsigned int pa;
+ unsigned long long pa_64;
+ };
+ union {
+ void *va;
+ unsigned long long va_padding;
+ };
+ union {
+ unsigned int size;
+ unsigned long long size_padding;
+ };
+};
+
+/**
+ * struct venc_work_buf_list - Structure for encode working buffer list
+ * @rc_code: RC code buffer
+ * @rc_info: RC info buffer
+ * @luma: luma buffer
+ * @chroma: chroma buffer
+ * @sub_luma: sub luma buffer
+ * @sub_write: sub write buffer
+ * @col_mv: col_mv buffer
+ * @wpp: wpp buffer
+ * @wpp_nbm: wpp nbm buffer
+ * @skip_frame: skip frame buffer
+ */
+struct venc_work_buf_list {
+ struct venc_work_buf rc_code;
+ struct venc_work_buf rc_info[MAX_VENC_CORE];
+ struct venc_work_buf luma[MAX_DPB_SIZE];
+ struct venc_work_buf chroma[MAX_DPB_SIZE];
+ struct venc_work_buf sub_luma[MAX_DPB_SIZE];
+ struct venc_work_buf sub_write[MAX_DPB_SIZE];
+ struct venc_work_buf col_mv[MAX_DPB_SIZE];
+ struct venc_work_buf wpp[MAX_VENC_CORE];
+ struct venc_work_buf wpp_nbm[MAX_VENC_CORE];
+ struct venc_work_buf skip_frame;
+};
+
+/**
+ * struct venc_info - Structure for encode frame and bs information
+ * @fb_addr: frame buffer address array
+ * @fb_size: frame buffer size array
+ * @bs_addr: bitstream buffer address
+ * @bs_size: bitstream buffer size
+ */
+struct venc_info {
+ unsigned long long fb_addr[VIDEO_MAX_PLANES];
+ unsigned int fb_size[VIDEO_MAX_PLANES];
+ unsigned long long bs_addr;
+ unsigned long long bs_size;
+};
+
+/**
+ * struct venc_vsi - Structure for VCP driver control and info share
+ * AP-W/R : AP is writer/reader on this item
+ * VCP-W/R: VCP is write/reader on this item
+ * @config: encoder configuration
+ * @data: encoder configuration data
+ * @bufs: encoder working buffers
+ * @venc: encoder information
+ */
+struct venc_vsi {
+ struct venc_config config;
+ struct venc_config_data data;
+ struct venc_work_buf_list bufs;
+ struct venc_info venc;
+};
+
+/**
+ * struct venc_inst - Structure for encoder instance
+ * @hw_base: hardware io address
+ * @pps_buf: PPS buffer
+ * @seq_buf: sequence header buffer
+ * @work_buf_allocated: work buffer allocated or not
+ * @frm_cnt: encoded frame count
+ * @skip_frm_cnt: encoded skip frame count
+ * @prepend_hdr: prepend header flag
+ * @vpu_inst: vpu instance
+ * @vsi: encode vsi
+ * @ctx: encoder context
+ */
+struct venc_inst {
+ void __iomem *hw_base;
+ struct mtk_vcodec_mem pps_buf;
+ struct mtk_vcodec_mem seq_buf;
+ bool work_buf_allocated;
+ unsigned int frm_cnt;
+ unsigned int skip_frm_cnt;
+ unsigned int prepend_hdr;
+ struct venc_vpu_inst vpu_inst;
+ struct venc_vsi *vsi;
+ struct mtk_vcodec_enc_ctx *ctx;
+};
+
+static int venc_init(struct mtk_vcodec_enc_ctx *ctx)
+{
+ int ret = 0;
+ struct venc_inst *inst;
+
+ inst = kzalloc_obj(inst, GFP_KERNEL);
+ if (!inst)
+ return -ENOMEM;
+
+ inst->ctx = ctx;
+ inst->vpu_inst.ctx = ctx;
+ inst->vpu_inst.id = mtk_vcodec_fw_get_venc_ipi(ctx->dev->fw_handler->type);
+ inst->hw_base = mtk_vcodec_get_reg_addr(inst->ctx->dev->reg_base, VENC_SYS);
+
+ ret = vpu_enc_init(&inst->vpu_inst);
+ inst->vsi = (struct venc_vsi *)inst->vpu_inst.vsi;
+
+ if (ret) {
+ kfree(inst);
+ return ret;
+ }
+
+ ctx->drv_handle = inst;
+
+ return 0;
+}
+
+static inline u32 venc_read_reg(struct venc_inst *inst, u32 addr)
+{
+ return readl(inst->hw_base + addr);
+}
+
+static unsigned int venc_wait_encode_done(struct venc_inst *inst)
+{
+ unsigned int irq_status = 0;
+ struct mtk_vcodec_enc_ctx *ctx = (struct mtk_vcodec_enc_ctx *)inst->ctx;
+
+ if (!mtk_vcodec_wait_for_done_ctx(ctx, MTK_INST_IRQ_RECEIVED,
+ WAIT_INTR_TIMEOUT_MS, 0)) {
+ irq_status = ctx->irq_status;
+ mtk_venc_debug(ctx, "irq_status %x <-", irq_status);
+ }
+ return irq_status;
+}
+
+static void venc_set_bufs(struct venc_inst *inst,
+ struct venc_frm_buf *frm_buf,
+ struct mtk_vcodec_mem *bs_buf)
+{
+ unsigned int i;
+
+ if (frm_buf) {
+ for (i = 0; i < frm_buf->num_planes; i++) {
+ inst->vsi->venc.fb_addr[i] = frm_buf->fb_addr[i].dma_addr;
+ inst->vsi->venc.fb_size[i] = frm_buf->fb_addr[i].size;
+ mtk_venc_debug(inst->ctx, "%s: fb_buf[%d]: %llx(%d)\n",
+ __func__, i,
+ inst->vsi->venc.fb_addr[i],
+ inst->vsi->venc.fb_size[i]);
+ }
+ }
+
+ if (bs_buf) {
+ inst->vsi->venc.bs_addr = bs_buf->dma_addr;
+ inst->vsi->venc.bs_size = bs_buf->size;
+ mtk_venc_debug(inst->ctx, "%s: bs_buf: %llx(%d)\n",
+ __func__,
+ inst->vsi->venc.bs_addr,
+ (unsigned int)inst->vsi->venc.bs_size);
+ }
+}
+
+static int venc_encode_sps(struct venc_inst *inst,
+ struct mtk_vcodec_mem *bs_buf,
+ unsigned int *bs_size)
+{
+ int ret = 0;
+ unsigned int irq_status;
+
+ venc_set_bufs(inst, NULL, bs_buf);
+ ret = vpu_enc_encode(&inst->vpu_inst, VENC_BS_MODE_SPS, NULL, bs_buf, NULL);
+ if (ret)
+ return ret;
+
+ irq_status = venc_wait_encode_done(inst);
+ if (irq_status != MTK_VENC_IRQ_STATUS_SPS) {
+ mtk_venc_err(inst->ctx, "expect irq status %d", MTK_VENC_IRQ_STATUS_SPS);
+ return -EINVAL;
+ }
+
+ *bs_size = venc_read_reg(inst, VENC_PIC_BITSTREAM_BYTE_CNT);
+ mtk_venc_debug(inst->ctx, "sps bs size %d <-", *bs_size);
+
+ return ret;
+}
+
+static int venc_encode_pps(struct venc_inst *inst,
+ struct mtk_vcodec_mem *bs_buf,
+ unsigned int *bs_size)
+{
+ int ret = 0;
+ unsigned int irq_status;
+
+ venc_set_bufs(inst, NULL, bs_buf);
+ ret = vpu_enc_encode(&inst->vpu_inst, VENC_BS_MODE_PPS, NULL, bs_buf, NULL);
+ if (ret)
+ return ret;
+
+ irq_status = venc_wait_encode_done(inst);
+ if (irq_status != MTK_VENC_IRQ_STATUS_PPS) {
+ mtk_venc_err(inst->ctx, "expect irq status %d", MTK_VENC_IRQ_STATUS_PPS);
+ return -EINVAL;
+ }
+
+ *bs_size = venc_read_reg(inst, VENC_PIC_BITSTREAM_BYTE_CNT);
+ mtk_venc_debug(inst->ctx, "pps bs size %d <-", *bs_size);
+
+ return ret;
+}
+
+static int venc_encode_header(struct venc_inst *inst,
+ struct mtk_vcodec_mem *bs_buf,
+ unsigned int *bs_size)
+{
+ int ret = 0;
+ unsigned int bs_size_sps;
+ unsigned int bs_size_pps;
+
+ ret = venc_encode_sps(inst, bs_buf, &bs_size_sps);
+ if (ret)
+ return ret;
+
+ ret = venc_encode_pps(inst, &inst->pps_buf, &bs_size_pps);
+ if (ret)
+ return ret;
+
+ memcpy(bs_buf->va + bs_size_sps, inst->pps_buf.va, bs_size_pps);
+ *bs_size = bs_size_sps + bs_size_pps;
+
+ return ret;
+}
+
+static int venc_encode_frame(struct venc_inst *inst,
+ struct venc_frm_buf *frm_buf,
+ struct mtk_vcodec_mem *bs_buf,
+ unsigned int *bs_size)
+{
+ int ret = 0;
+ unsigned int irq_status;
+
+ venc_set_bufs(inst, frm_buf, bs_buf);
+ ret = vpu_enc_encode(&inst->vpu_inst, VENC_BS_MODE_FRAME, frm_buf, bs_buf, NULL);
+ if (ret)
+ return ret;
+
+ irq_status = venc_wait_encode_done(inst);
+ if (irq_status != MTK_VENC_IRQ_STATUS_FRM) {
+ mtk_venc_err(inst->ctx, "expect irq status %d", MTK_VENC_IRQ_STATUS_FRM);
+ return -EINVAL;
+ }
+
+ *bs_size = venc_read_reg(inst, VENC_PIC_BITSTREAM_BYTE_CNT);
+
+ ++inst->frm_cnt;
+
+ return ret;
+}
+
+static int venc_encode(void *handle,
+ enum venc_start_opt opt,
+ struct venc_frm_buf *frm_buf,
+ struct mtk_vcodec_mem *bs_buf,
+ struct venc_done_result *result)
+{
+ int ret = 0;
+ struct venc_inst *inst = (struct venc_inst *)handle;
+ struct mtk_vcodec_enc_ctx *ctx;
+ unsigned int bs_size_hdr;
+
+ if (WARN_ON(!inst || !inst->vsi))
+ return -EINVAL;
+
+ ctx = inst->ctx;
+
+ mtk_venc_debug(ctx, "%s: opt: %d\n", __func__, opt);
+
+ enable_irq(ctx->dev->enc_irq);
+ switch (opt) {
+ case VENC_START_OPT_ENCODE_SEQUENCE_HEADER: {
+ ret = venc_encode_header(inst, bs_buf, &bs_size_hdr);
+ if (ret)
+ goto encode_err;
+
+ result->bs_size = bs_size_hdr;
+ result->is_key_frm = false;
+ break;
+ }
+
+ case VENC_START_OPT_ENCODE_FRAME: {
+ if (!inst->prepend_hdr) {
+ ret = venc_encode_frame(inst, frm_buf, bs_buf, &result->bs_size);
+ if (ret)
+ goto encode_err;
+
+ result->is_key_frm = inst->vpu_inst.is_key_frm;
+ break;
+ }
+
+ ret = venc_encode_header(inst, &inst->seq_buf, &bs_size_hdr);
+ if (ret)
+ goto encode_err;
+
+ ret = venc_encode_frame(inst, frm_buf, bs_buf, &result->bs_size);
+ if (ret)
+ goto encode_err;
+
+ memmove(bs_buf->va + bs_size_hdr, bs_buf->va, result->bs_size);
+ memcpy(bs_buf->va, inst->seq_buf.va, bs_size_hdr);
+ result->bs_size += bs_size_hdr;
+
+ inst->prepend_hdr = 0;
+ result->is_key_frm = inst->vpu_inst.is_key_frm;
+ break;
+ }
+
+ default:
+ mtk_venc_err(inst->ctx, "venc_opt %d not supported", opt);
+ ret = -EINVAL;
+ break;
+ }
+
+encode_err:
+ disable_irq(ctx->dev->enc_irq);
+ mtk_venc_debug(ctx, "opt %d, return %d", opt, ret);
+
+ return ret;
+}
+
+static int mtk_venc_mem_alloc(struct venc_inst *inst,
+ struct device *dev,
+ struct venc_work_buf *buf)
+{
+ dma_addr_t dma_addr;
+
+ if (WARN_ON(!dev || !buf))
+ return -EINVAL;
+
+ if (buf->size == 0)
+ return 0;
+
+ buf->va = dma_alloc_coherent(dev, buf->size, &dma_addr, GFP_KERNEL);
+ if (!buf->va)
+ return -ENOMEM;
+
+ buf->iova = (unsigned long long)dma_addr;
+
+ mtk_venc_debug(inst->ctx, "allocate buffer, size: %d, va: %p, iova: 0x%llx",
+ buf->size, buf->va, buf->iova);
+
+ return 0;
+}
+
+static void mtk_venc_mem_free(struct venc_inst *inst,
+ struct device *dev,
+ struct venc_work_buf *buf)
+{
+ if (WARN_ON(!dev || !buf))
+ return;
+
+ if (!buf->va)
+ return;
+
+ mtk_venc_debug(inst->ctx, "free buffer, size: %d, va: %p, iova: 0x%llx",
+ buf->size, buf->va, buf->iova);
+
+ dma_free_coherent(dev, buf->size, buf->va, buf->iova);
+ buf->va = NULL;
+ buf->iova = 0;
+ buf->size = 0;
+}
+
+static void venc_free_rc_buf(struct venc_inst *inst,
+ struct venc_work_buf_list *bufs,
+ unsigned int core_num)
+{
+ int i;
+ struct device *dev;
+
+ dev = &inst->ctx->dev->plat_dev->dev;
+ mtk_venc_mem_free(inst, dev, &bufs->rc_code);
+
+ for (i = 0; i < core_num; i++)
+ mtk_venc_mem_free(inst, dev, &bufs->rc_info[i]);
+}
+
+static void venc_free_work_buf(struct venc_inst *inst)
+{
+ int i;
+ struct venc_work_buf_list *bufs = &inst->vsi->bufs;
+ unsigned int core_num = inst->vsi->config.core_num;
+ unsigned int dpb_size = inst->vsi->config.dpb_size;
+ struct device *dev;
+
+ if (bufs->rc_code.va)
+ venc_free_rc_buf(inst, bufs, core_num);
+
+ dev = &inst->ctx->dev->plat_dev->dev;
+
+ for (i = 0; i < core_num; i++) {
+ mtk_venc_mem_free(inst, dev, &bufs->wpp[i]);
+ mtk_venc_mem_free(inst, dev, &bufs->wpp_nbm[i]);
+ }
+
+ for (i = 0; i < dpb_size; i++) {
+ mtk_venc_mem_free(inst, dev, &bufs->luma[i]);
+ mtk_venc_mem_free(inst, dev, &bufs->chroma[i]);
+ mtk_venc_mem_free(inst, dev, &bufs->sub_luma[i]);
+ mtk_venc_mem_free(inst, dev, &bufs->sub_write[i]);
+ mtk_venc_mem_free(inst, dev, &bufs->col_mv[i]);
+ }
+
+ if (inst->pps_buf.va)
+ mtk_vcodec_mem_free(inst->ctx, &inst->pps_buf);
+
+ if (inst->seq_buf.va)
+ mtk_vcodec_mem_free(inst->ctx, &inst->seq_buf);
+}
+
+static int venc_alloc_rc_buf(struct venc_inst *inst,
+ struct venc_work_buf_list *bufs,
+ unsigned int core_num)
+{
+ int i;
+ struct mtk_vcodec_fw *fw = inst->ctx->dev->fw_handler;
+ struct device *dev;
+ void *tmp_va;
+
+ dev = &inst->ctx->dev->plat_dev->dev;
+ if (mtk_venc_mem_alloc(inst, dev, &bufs->rc_code))
+ return -ENOMEM;
+
+ tmp_va = mtk_vcodec_fw_map_dm_addr(fw, bufs->rc_code.pa);
+ memcpy(bufs->rc_code.va, tmp_va, bufs->rc_code.size);
+
+ for (i = 0; i < core_num; i++) {
+ if (mtk_venc_mem_alloc(inst, dev, &bufs->rc_info[i]))
+ goto err_rc_buf;
+ }
+
+ return 0;
+
+err_rc_buf:
+ venc_free_rc_buf(inst, bufs, core_num);
+
+ return -ENOMEM;
+}
+
+static int venc_alloc_work_buf(struct venc_inst *inst)
+{
+ int i, ret;
+ struct venc_work_buf_list *bufs = &inst->vsi->bufs;
+ unsigned int core_num = inst->vsi->config.core_num;
+ unsigned int dpb_size = inst->vsi->config.dpb_size;
+ struct device *dev;
+
+ if (bufs->rc_code.size != 0) {
+ ret = venc_alloc_rc_buf(inst, bufs, core_num);
+ if (ret) {
+ mtk_venc_err(inst->ctx, "cannot allocate rc buf");
+ return -ENOMEM;
+ }
+ }
+
+ dev = &inst->ctx->dev->plat_dev->dev;
+
+ for (i = 0; i < core_num; i++) {
+ if (mtk_venc_mem_alloc(inst, dev, &bufs->wpp[i]) ||
+ mtk_venc_mem_alloc(inst, dev, &bufs->wpp_nbm[i]))
+ goto err_alloc;
+ }
+
+ for (i = 0; i < dpb_size; i++) {
+ if (mtk_venc_mem_alloc(inst, dev, &bufs->luma[i]) ||
+ mtk_venc_mem_alloc(inst, dev, &bufs->chroma[i]) ||
+ mtk_venc_mem_alloc(inst, dev, &bufs->sub_luma[i]) ||
+ mtk_venc_mem_alloc(inst, dev, &bufs->sub_write[i]) ||
+ mtk_venc_mem_alloc(inst, dev, &bufs->col_mv[i]))
+ goto err_alloc;
+ }
+
+ /* the pps_buf and seq_buf are used by AP side only */
+ inst->pps_buf.size = PPS_SIZE;
+ ret = mtk_vcodec_mem_alloc(inst->ctx, &inst->pps_buf);
+ if (ret) {
+ mtk_venc_err(inst->ctx, "cannot allocate pps_buf");
+ goto err_alloc;
+ }
+
+ inst->seq_buf.size = SEQ_HEADER_SIZE;
+ ret = mtk_vcodec_mem_alloc(inst->ctx, &inst->seq_buf);
+ if (ret) {
+ mtk_venc_err(inst->ctx, "cannot allocate seq_buf");
+ goto err_alloc;
+ }
+ return 0;
+
+err_alloc:
+ venc_free_work_buf(inst);
+ return -ENOMEM;
+}
+
+static int venc_set_param(void *handle,
+ enum venc_set_param_type type,
+ struct venc_enc_param *enc_prm)
+{
+ int ret = 0;
+ struct venc_inst *inst = (struct venc_inst *)handle;
+
+ switch (type) {
+ case VENC_SET_PARAM_ENC:
+ if (WARN_ON(!inst->vsi))
+ return -EINVAL;
+ inst->vsi->config.input_fourcc = enc_prm->input_yuv_fmt;
+ inst->vsi->config.bitrate = enc_prm->bitrate;
+ inst->vsi->config.pic_w = enc_prm->width;
+ inst->vsi->config.pic_h = enc_prm->height;
+ inst->vsi->config.buf_w = enc_prm->buf_width;
+ inst->vsi->config.buf_h = enc_prm->buf_height;
+ inst->vsi->config.gop_size = enc_prm->gop_size;
+ inst->vsi->config.framerate = enc_prm->frm_rate;
+ inst->vsi->config.intra_period = enc_prm->intra_period;
+ inst->vsi->config.profile = enc_prm->profile;
+ inst->vsi->config.level = enc_prm->level;
+
+ ret = vpu_enc_set_param(&inst->vpu_inst, type, enc_prm);
+ if (ret)
+ break;
+
+ if (inst->work_buf_allocated) {
+ venc_free_work_buf(inst);
+ inst->work_buf_allocated = false;
+ }
+ ret = venc_alloc_work_buf(inst);
+ if (ret)
+ break;
+ inst->work_buf_allocated = true;
+ break;
+ case VENC_SET_PARAM_PREPEND_HEADER:
+ inst->prepend_hdr = 1;
+ break;
+ default:
+ ret = vpu_enc_set_param(&inst->vpu_inst, type, enc_prm);
+ break;
+ }
+
+ return ret;
+}
+
+static int venc_deinit(void *handle)
+{
+ int ret = 0;
+ struct venc_inst *inst = (struct venc_inst *)handle;
+
+ ret = vpu_enc_deinit(&inst->vpu_inst);
+
+ if (inst->work_buf_allocated)
+ venc_free_work_buf(inst);
+
+ kfree(inst);
+
+ return ret;
+}
+
+const struct venc_common_if venc_if = {
+ .init = venc_init,
+ .encode = venc_encode,
+ .set_param = venc_set_param,
+ .deinit = venc_deinit,
+};
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_h264_if.c b/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_h264_if.c
index d2f4d732d2f7..320c505cdb21 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_h264_if.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/venc/venc_h264_if.c
@@ -723,9 +723,9 @@ static void h264_enc_set_vsi_configs(struct venc_h264_inst *inst,
inst->vsi->config.framerate = enc_prm->frm_rate;
inst->vsi->config.intra_period = enc_prm->intra_period;
inst->vsi->config.profile =
- h264_get_profile(inst, enc_prm->h264_profile);
+ h264_get_profile(inst, enc_prm->profile);
inst->vsi->config.level =
- h264_get_level(inst, enc_prm->h264_level);
+ h264_get_level(inst, enc_prm->level);
inst->vsi->config.wfd = 0;
}
@@ -742,9 +742,9 @@ static void h264_enc_set_vsi_34_configs(struct venc_h264_inst *inst,
inst->vsi_34->config.framerate = enc_prm->frm_rate;
inst->vsi_34->config.intra_period = enc_prm->intra_period;
inst->vsi_34->config.profile =
- h264_get_profile(inst, enc_prm->h264_profile);
+ h264_get_profile(inst, enc_prm->profile);
inst->vsi_34->config.level =
- h264_get_level(inst, enc_prm->h264_level);
+ h264_get_level(inst, enc_prm->level);
inst->vsi_34->config.wfd = 0;
}
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.h b/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.h
index 889440a436b6..3c2a1b5e9312 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.h
+++ b/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.h
@@ -66,8 +66,8 @@ enum venc_set_param_type {
* struct venc_enc_prm - encoder settings for VENC_SET_PARAM_ENC used in
* venc_if_set_param()
* @input_fourcc: input yuv format
- * @h264_profile: V4L2 defined H.264 profile
- * @h264_level: V4L2 defined H.264 level
+ * @profile: V4L2 defined profile
+ * @level: V4L2 defined level
* @width: image width
* @height: image height
* @buf_width: buffer width
@@ -79,8 +79,8 @@ enum venc_set_param_type {
*/
struct venc_enc_param {
enum venc_yuv_fmt input_yuv_fmt;
- unsigned int h264_profile;
- unsigned int h264_level;
+ unsigned int profile;
+ unsigned int level;
unsigned int width;
unsigned int height;
unsigned int buf_width;
@@ -107,9 +107,11 @@ struct venc_frame_info {
/*
* struct venc_frm_buf - frame buffer information used in venc_if_encode()
* @fb_addr: plane frame buffer addresses
+ * @num_planes: number of planes
*/
struct venc_frm_buf {
struct mtk_vcodec_fb fb_addr[MTK_VCODEC_MAX_PLANES];
+ unsigned int num_planes;
};
/*
@@ -124,6 +126,7 @@ struct venc_done_result {
extern const struct venc_common_if venc_h264_if;
extern const struct venc_common_if venc_vp8_if;
+extern const struct venc_common_if venc_if;
/*
* venc_if_init - Create the driver handle
--
2.45.2
^ permalink raw reply related
* [PATCH v6 6/6] media: mediatek: encoder: Add MT8196 encoder compatible data
From: Irui Wang @ 2026-04-23 7:33 UTC (permalink / raw)
To: Hans Verkuil, Mauro Carvalho Chehab, Rob Herring,
Matthias Brugger, Krzysztof Kozlowski, angelogioacchino.delregno,
nicolas.dufresne, Tiffany Lin, kyrie wu
Cc: Yunfei Dong, Maoguang Meng, Longfei Wang, Irui Wang,
Project_Global_Chrome_Upstream_Group, linux-media, devicetree,
linux-kernel, linux-arm-kernel, linux-mediatek
In-Reply-To: <20260423073345.27402-1-irui.wang@mediatek.com>
MT8196 encoder use common firmware interface, add compatible data to
support MT8196 encoding, and need set dma mask to support 34bit.
Signed-off-by: Irui Wang <irui.wang@mediatek.com>
---
.../vcodec/encoder/mtk_vcodec_enc_drv.c | 19 +++++++++++++++++++
.../vcodec/encoder/mtk_vcodec_enc_drv.h | 2 ++
2 files changed, 21 insertions(+)
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c
index 7b644f55963b..d7328d013ff6 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c
@@ -20,6 +20,8 @@
#include "mtk_vcodec_enc_pm.h"
#include "../common/mtk_vcodec_intr.h"
+#define VENC_DMA_BIT_MASK 34
+
static const struct mtk_video_fmt mtk_video_formats_output[] = {
{
.fourcc = V4L2_PIX_FMT_NV12M,
@@ -309,6 +311,9 @@ static int mtk_vcodec_probe(struct platform_device *pdev)
goto err_res;
}
+ if (dev->venc_pdata->set_dma_bit_mask)
+ dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(VENC_DMA_BIT_MASK));
+
mutex_init(&dev->enc_mutex);
mutex_init(&dev->dev_mutex);
spin_lock_init(&dev->dev_ctx_lock);
@@ -460,6 +465,19 @@ static const struct mtk_vcodec_enc_pdata mt8195_pdata = {
.core_id = VENC_SYS,
};
+static const struct mtk_vcodec_enc_pdata mt8196_pdata = {
+ .venc_model_num = 8196,
+ .capture_formats = mtk_video_formats_capture_h264,
+ .num_capture_formats = ARRAY_SIZE(mtk_video_formats_capture_h264),
+ .output_formats = mtk_video_formats_output,
+ .num_output_formats = ARRAY_SIZE(mtk_video_formats_output),
+ .min_bitrate = 64,
+ .max_bitrate = 100000000,
+ .core_id = VENC_SYS,
+ .uses_common_fw_iface = true,
+ .set_dma_bit_mask = true,
+};
+
static const struct of_device_id mtk_vcodec_enc_match[] = {
{.compatible = "mediatek,mt8173-vcodec-enc",
.data = &mt8173_avc_pdata},
@@ -469,6 +487,7 @@ static const struct of_device_id mtk_vcodec_enc_match[] = {
{.compatible = "mediatek,mt8188-vcodec-enc", .data = &mt8188_pdata},
{.compatible = "mediatek,mt8192-vcodec-enc", .data = &mt8192_pdata},
{.compatible = "mediatek,mt8195-vcodec-enc", .data = &mt8195_pdata},
+ {.compatible = "mediatek,mt8196-vcodec-enc", .data = &mt8196_pdata},
{},
};
MODULE_DEVICE_TABLE(of, mtk_vcodec_enc_match);
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
index 769fb5009964..475953d39aa4 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
@@ -32,6 +32,7 @@
* @core_id: stand for h264 or vp8 encode index
* @uses_34bit: whether the encoder uses 34-bit iova
* @uses_common_fw_iface: whether the encoder uses common driver interface
+ * @set_dma_bit_mask: whether the encoder need set extra DMA bit mask
*/
struct mtk_vcodec_enc_pdata {
u16 venc_model_num;
@@ -45,6 +46,7 @@ struct mtk_vcodec_enc_pdata {
u8 core_id;
bool uses_34bit;
bool uses_common_fw_iface;
+ bool set_dma_bit_mask;
};
/*
--
2.45.2
^ permalink raw reply related
* [PATCH v6 3/6] media: mediatek: encoder: Add support for common firmware interface
From: Irui Wang @ 2026-04-23 7:33 UTC (permalink / raw)
To: Hans Verkuil, Mauro Carvalho Chehab, Rob Herring,
Matthias Brugger, Krzysztof Kozlowski, angelogioacchino.delregno,
nicolas.dufresne, Tiffany Lin, kyrie wu
Cc: Yunfei Dong, Maoguang Meng, Longfei Wang, Irui Wang,
Project_Global_Chrome_Upstream_Group, linux-media, devicetree,
linux-kernel, linux-arm-kernel, linux-mediatek
In-Reply-To: <20260423073345.27402-1-irui.wang@mediatek.com>
The existing encoder firmware interface implied just one type of codec:
H.264. Future encoders may support additional codecs; however adding
entire sets of interfaces for them is not scalable.
Instead, a new "common" firmware interface is defined for non codec
specific messages. The new messages encapsulate the old ones for
backward compatibility.
This patch adds support for these new messages.
Signed-off-by: Irui Wang <irui.wang@mediatek.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
---
.../vcodec/encoder/mtk_vcodec_enc_drv.h | 3 ++
.../mediatek/vcodec/encoder/venc_drv_if.c | 3 +-
.../mediatek/vcodec/encoder/venc_ipi_msg.h | 26 +++++++++++++++
.../mediatek/vcodec/encoder/venc_vpu_if.c | 33 ++++++++++++-------
4 files changed, 52 insertions(+), 13 deletions(-)
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
index 53369adc083b..0529564027c4 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
+++ b/drivers/media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.h
@@ -16,6 +16,7 @@
#define MTK_ENC_CTX_IS_EXT(ctx) ((ctx)->dev->venc_pdata->uses_ext)
#define MTK_ENC_IOVA_IS_34BIT(ctx) ((ctx)->dev->venc_pdata->uses_34bit)
+#define MTK_ENC_DRV_IS_COMM(ctx) (((ctx)->dev->venc_pdata->uses_common_fw_iface))
/**
* struct mtk_vcodec_enc_pdata - compatible data for each IC
@@ -29,6 +30,7 @@
* @num_output_formats: number of entries in output_formats
* @core_id: stand for h264 or vp8 encode index
* @uses_34bit: whether the encoder uses 34-bit iova
+ * @uses_common_fw_iface: whether the encoder uses common driver interface
*/
struct mtk_vcodec_enc_pdata {
bool uses_ext;
@@ -40,6 +42,7 @@ struct mtk_vcodec_enc_pdata {
size_t num_output_formats;
u8 core_id;
bool uses_34bit;
+ bool uses_common_fw_iface;
};
/*
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.c b/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.c
index e83747b8d69a..f8c9349c18c0 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/venc_drv_if.c
@@ -19,13 +19,14 @@
int venc_if_init(struct mtk_vcodec_enc_ctx *ctx, unsigned int fourcc)
{
int ret = 0;
+ const bool uses_common_fw_iface = MTK_ENC_DRV_IS_COMM(ctx);
switch (fourcc) {
case V4L2_PIX_FMT_VP8:
ctx->enc_if = &venc_vp8_if;
break;
case V4L2_PIX_FMT_H264:
- ctx->enc_if = &venc_h264_if;
+ ctx->enc_if = uses_common_fw_iface ? &venc_if : &venc_h264_if;
break;
default:
return -EINVAL;
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/venc_ipi_msg.h b/drivers/media/platform/mediatek/vcodec/encoder/venc_ipi_msg.h
index bb16d96a7f57..ce3c2c8059fb 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/venc_ipi_msg.h
+++ b/drivers/media/platform/mediatek/vcodec/encoder/venc_ipi_msg.h
@@ -45,6 +45,20 @@ struct venc_ap_ipi_msg_init {
uint64_t venc_inst;
};
+/**
+ * struct venc_ap_ipi_msg_init_comm - AP to VPU init cmd structure
+ * @base: AP to VPU init cmd structure
+ * @codec_type: encoder type
+ * @reserved: reserved field
+ * @shared_iova: shared iova
+ */
+struct venc_ap_ipi_msg_init_comm {
+ struct venc_ap_ipi_msg_init base;
+ u32 codec_type;
+ u32 reserved;
+ u64 shared_iova;
+};
+
/**
* struct venc_ap_ipi_msg_set_param - AP to VPU set_param cmd structure
* @msg_id: message id (AP_IPIMSG_XXX_ENC_SET_PARAM)
@@ -175,6 +189,18 @@ struct venc_vpu_ipi_msg_init {
uint32_t venc_abi_version;
};
+/**
+ * struct venc_vpu_ipi_msg_init_comm - VPU ack AP init cmd structure
+ * @init_ack: AP init cmd structure
+ * @vpu_vsi_addr: VSI address from VPU
+ * @reserved: reserved field
+ */
+struct venc_vpu_ipi_msg_init_comm {
+ struct venc_vpu_ipi_msg_init init_ack;
+ u32 vpu_vsi_addr;
+ u32 reserved;
+};
+
/**
* struct venc_vpu_ipi_msg_set_param - VPU ack AP set_param cmd structure
* @msg_id: message id (VPU_IPIMSG_XXX_ENC_SET_PARAM_DONE)
diff --git a/drivers/media/platform/mediatek/vcodec/encoder/venc_vpu_if.c b/drivers/media/platform/mediatek/vcodec/encoder/venc_vpu_if.c
index 0c825aa7224d..7772b8442ebc 100644
--- a/drivers/media/platform/mediatek/vcodec/encoder/venc_vpu_if.c
+++ b/drivers/media/platform/mediatek/vcodec/encoder/venc_vpu_if.c
@@ -10,24 +10,25 @@
static void handle_enc_init_msg(struct venc_vpu_inst *vpu, const void *data)
{
- const struct venc_vpu_ipi_msg_init *msg = data;
+ const struct venc_vpu_ipi_msg_init_comm *msg = data;
+ struct mtk_vcodec_fw *fw = vpu->ctx->dev->fw_handler;
- vpu->inst_addr = msg->vpu_inst_addr;
- vpu->vsi = mtk_vcodec_fw_map_dm_addr(vpu->ctx->dev->fw_handler,
- msg->vpu_inst_addr);
+ vpu->inst_addr = msg->init_ack.vpu_inst_addr;
+ vpu->vsi = mtk_vcodec_fw_map_dm_addr(fw, vpu->inst_addr);
/* Firmware version field value is unspecified on MT8173. */
- if (mtk_vcodec_fw_get_type(vpu->ctx->dev->fw_handler) == VPU)
+ if (mtk_vcodec_fw_get_type(fw) == VPU)
return;
/* Check firmware version. */
- mtk_venc_debug(vpu->ctx, "firmware version: 0x%x\n", msg->venc_abi_version);
- switch (msg->venc_abi_version) {
+ mtk_venc_debug(vpu->ctx, "firmware version: 0x%x\n",
+ msg->init_ack.venc_abi_version);
+ switch (msg->init_ack.venc_abi_version) {
case 1:
break;
default:
mtk_venc_err(vpu->ctx, "unhandled firmware version 0x%x\n",
- msg->venc_abi_version);
+ msg->init_ack.venc_abi_version);
vpu->failure = 1;
break;
}
@@ -133,7 +134,8 @@ static int vpu_enc_send_msg(struct venc_vpu_inst *vpu, void *msg,
int vpu_enc_init(struct venc_vpu_inst *vpu)
{
int status;
- struct venc_ap_ipi_msg_init out = { };
+ size_t msg_size;
+ struct venc_ap_ipi_msg_init_comm out = { };
init_waitqueue_head(&vpu->wq_hd);
vpu->signaled = 0;
@@ -149,9 +151,16 @@ int vpu_enc_init(struct venc_vpu_inst *vpu)
return -EINVAL;
}
- out.msg_id = AP_IPIMSG_ENC_INIT;
- out.venc_inst = (unsigned long)vpu;
- if (vpu_enc_send_msg(vpu, &out, sizeof(out))) {
+ out.base.msg_id = AP_IPIMSG_ENC_INIT;
+ out.base.venc_inst = (unsigned long)vpu;
+ if (MTK_ENC_DRV_IS_COMM(vpu->ctx)) {
+ out.codec_type = vpu->ctx->q_data[MTK_Q_DATA_DST].fmt->fourcc;
+ msg_size = sizeof(struct venc_ap_ipi_msg_init_comm);
+ } else {
+ msg_size = sizeof(struct venc_ap_ipi_msg_init);
+ }
+
+ if (vpu_enc_send_msg(vpu, &out, msg_size)) {
mtk_venc_err(vpu->ctx, "AP_IPIMSG_ENC_INIT fail");
return -EINVAL;
}
--
2.45.2
^ permalink raw reply related
* Re: [PATCH net] net: airoha: stop net_device TX queue before updating CPU index
From: Lorenzo Bianconi @ 2026-04-23 7:42 UTC (permalink / raw)
To: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni
Cc: Simon Horman, linux-arm-kernel, linux-mediatek, netdev
In-Reply-To: <20260421-airoha-xmit-stop-condition-v1-1-e670d6a48467@kernel.org>
[-- Attachment #1: Type: text/plain, Size: 2890 bytes --]
> Currently, airoha_eth driver updates the CPU index register prior of
> verifying whether the number of free descriptors has fallen below the
> threshold.
> Move net_device TX queue length check before updating the TX CPU index
> in order to update TX CPU index even if there are more packets to be
> transmitted but the net_device TX queue is going to be stopped
> accounting the inflight packets.
>
> Fixes: 1d304174106c ("net: airoha: Implement BQL support")
> Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
> ---
> drivers/net/ethernet/airoha/airoha_eth.c | 9 ++++-----
> 1 file changed, 4 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c
> index 19f67c7dd8e1..5d327237e274 100644
> --- a/drivers/net/ethernet/airoha/airoha_eth.c
> +++ b/drivers/net/ethernet/airoha/airoha_eth.c
> @@ -2058,17 +2058,16 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb,
>
> skb_tx_timestamp(skb);
> netdev_tx_sent_queue(txq, skb->len);
> + if (q->ndesc - q->queued < q->free_thr) {
> + netif_tx_stop_queue(txq);
> + q->txq_stopped = true;
> + }
>
> if (netif_xmit_stopped(txq) || !netdev_xmit_more())
> airoha_qdma_rmw(qdma, REG_TX_CPU_IDX(qid),
> TX_RING_CPU_IDX_MASK,
> FIELD_PREP(TX_RING_CPU_IDX_MASK, index));
>
> - if (q->ndesc - q->queued < q->free_thr) {
> - netif_tx_stop_queue(txq);
> - q->txq_stopped = true;
> - }
> -
> spin_unlock_bh(&q->lock);
>
> return NETDEV_TX_OK;
>
> ---
> base-commit: a663bac71a2f0b3ac6c373168ca57b2a6e6381aa
> change-id: 20260421-airoha-xmit-stop-condition-344dc0292a19
>
> Best regards,
> --
> Lorenzo Bianconi <lorenzo@kernel.org>
>
commenting on Sashiko retported issues:
https://sashiko.dev/#/patchset/20260421-airoha-xmit-stop-condition-v1-1-e670d6a48467%40kernel.org
- Could this cause a deadlock if exactly q->free_thr descriptors are free?
This does not seem a problem to me since, even if the netdev tx queue is
stopped as described in the report, the airoha_qdma_tx_napi_poll() will free
space in the queue and subsequent packets will update REG_TX_CPU_IDX register.
- Is it possible for this loop to read past the end of the frags array?
As pointed out by Sashiko, this issue is not introduced by this patch and I
will fix with a dedicated patch.
- Might this lead to memory corruption if the tcp header is not in the linear area?
This issue is not introduced by this patch and I will fix with a dedicated patch.
- If an error occurs during transmission, the driver jumps to the error label
frees the skb, and returns NETDEV_TX_OK without ringing the qdma cpu index doorbell?
Similar to the first issue, this does not seem a problem to me since subsequent
packets will update REG_TX_CPU_IDX register.
Regards,
Lorenzo
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]
^ permalink raw reply
* Re: [PATCH v3 1/3] crypto: atmel-sha204a - fix memory leak at non-blocking RNG work_data
From: Ard Biesheuvel @ 2026-04-23 7:43 UTC (permalink / raw)
To: Lothar Rubusch, Herbert Xu, Thorsten Blum, davem, nicolas.ferre,
alexandre.belloni, claudiu.beznea, Linus Walleij
Cc: linux-crypto, linux-arm-kernel, linux-kernel
In-Reply-To: <20260422210936.20095-2-l.rubusch@gmail.com>
Hi Lothar,
On Wed, 22 Apr 2026, at 23:09, Lothar Rubusch wrote:
> The driver allocated memory for work_data in the non-blocking read
> path but never free'd it again. After first read-out the memory pointer
> seemed to be recycled and never was allocated again, due to some errors
> in the logic, so that the leak was not growing.
>
Why can't we just reuse the work_data, instead of alloc/freeing it every time?
> Add kfree(work_data) in the completion callback on error. then add
> kfree(work_data) after the data is consumed in the subsequent read
> call. Finally ensure atomic_dec() is called only after the data has
> been consumed or an error occurred to prevent race conditions.
>
> Fixes: da001fb651b0 ("crypto: atmel-i2c - add support for SHA204A
> random number generator")
> Signed-off-by: Lothar Rubusch <l.rubusch@gmail.com>
> ---
> drivers/crypto/atmel-sha204a.c | 43 ++++++++++++++++++++--------------
> 1 file changed, 26 insertions(+), 17 deletions(-)
>
> diff --git a/drivers/crypto/atmel-sha204a.c
> b/drivers/crypto/atmel-sha204a.c
> index dbb39ed0cea1..19720bdd446d 100644
> --- a/drivers/crypto/atmel-sha204a.c
> +++ b/drivers/crypto/atmel-sha204a.c
> @@ -25,13 +25,17 @@ static void atmel_sha204a_rng_done(struct
> atmel_i2c_work_data *work_data,
> struct atmel_i2c_client_priv *i2c_priv = work_data->ctx;
> struct hwrng *rng = areq;
>
> - if (status)
> + if (status) {
> dev_warn_ratelimited(&i2c_priv->client->dev,
> "i2c transaction failed (%d)\n",
> status);
> + kfree(work_data);
> + rng->priv = 0;
> + atomic_dec(&i2c_priv->tfm_count);
> + return;
> + }
>
> rng->priv = (unsigned long)work_data;
> - atomic_dec(&i2c_priv->tfm_count);
> }
>
> static int atmel_sha204a_rng_read_nonblocking(struct hwrng *rng, void
> *data,
> @@ -42,31 +46,36 @@ static int
> atmel_sha204a_rng_read_nonblocking(struct hwrng *rng, void *data,
>
> i2c_priv = container_of(rng, struct atmel_i2c_client_priv, hwrng);
>
> - /* keep maximum 1 asynchronous read in flight at any time */
> - if (!atomic_add_unless(&i2c_priv->tfm_count, 1, 1))
> - return 0;
> -
> + /* Verify if data available from last run */
> if (rng->priv) {
> work_data = (struct atmel_i2c_work_data *)rng->priv;
> max = min(sizeof(work_data->cmd.data), max);
> memcpy(data, &work_data->cmd.data, max);
> - rng->priv = 0;
> - } else {
> - work_data = kmalloc_obj(*work_data, GFP_ATOMIC);
> - if (!work_data) {
> - atomic_dec(&i2c_priv->tfm_count);
> - return -ENOMEM;
> - }
> - work_data->ctx = i2c_priv;
> - work_data->client = i2c_priv->client;
>
> - max = 0;
> + /* Now, free memory */
> + kfree(work_data);
> + rng->priv = 0;
> + atomic_dec(&i2c_priv->tfm_count);
> + return max;
> }
>
> + /* When a request is still in-flight but not processed */
> + if (atomic_read(&i2c_priv->tfm_count) > 0)
> + return 0;
> +
> + /* Start a new request */
> + work_data = kmalloc_obj(*work_data, GFP_ATOMIC);
> + if (!work_data)
> + return -ENOMEM;
> +
> + atomic_inc(&i2c_priv->tfm_count);
> + work_data->ctx = i2c_priv;
> + work_data->client = i2c_priv->client;
> +
> atmel_i2c_init_random_cmd(&work_data->cmd);
> atmel_i2c_enqueue(work_data, atmel_sha204a_rng_done, rng);
>
> - return max;
> + return 0;
> }
>
> static int atmel_sha204a_rng_read(struct hwrng *rng, void *data, size_t max,
> --
> 2.53.0
^ permalink raw reply
* Re: [PATCH 2/8] xor/arm: Replace vectorized implementation with arm64's intrinsics
From: Christoph Hellwig @ 2026-04-23 7:44 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: linux-arm-kernel, linux-crypto, linux-raid, Ard Biesheuvel,
Christoph Hellwig, Russell King, Arnd Bergmann, Eric Biggers
In-Reply-To: <20260422171655.3437334-12-ardb+git@google.com>
Nice!
Acked-by: Christoph Hellwig <hch@lst.de>
^ permalink raw reply
* Re: [PATCH 3/8] xor/arm64: Use shared NEON intrinsics implementation from 32-bit ARM
From: Christoph Hellwig @ 2026-04-23 7:46 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: linux-arm-kernel, linux-crypto, linux-raid, Ard Biesheuvel,
Christoph Hellwig, Russell King, Arnd Bergmann, Eric Biggers
In-Reply-To: <20260422171655.3437334-13-ardb+git@google.com>
> +extern void __xor_eor3_2(unsigned long bytes, unsigned long * __restrict p1,
> + const unsigned long * __restrict p2);
Does the alias magic prevent this from being in a header? If so a comment
would be nice, otherwise moving it to a header would be even better.
^ permalink raw reply
* Re: [PATCH 7/8] lib/raid6: Include asm/neon-intrinsics.h rather than arm_neon.h
From: Christoph Hellwig @ 2026-04-23 7:47 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: linux-arm-kernel, linux-crypto, linux-raid, Ard Biesheuvel,
Christoph Hellwig, Russell King, Arnd Bergmann, Eric Biggers
In-Reply-To: <20260422171655.3437334-17-ardb+git@google.com>
On Wed, Apr 22, 2026 at 07:17:03PM +0200, Ard Biesheuvel wrote:
> From: Ard Biesheuvel <ardb@kernel.org>
>
> arm_neon.h is a compiler header which needs some scaffolding to work
> correctly in the linux context, and so it is better not to include it
> directly. Both ARM and arm64 now provide asm/neon-intrinsics.h which
> takes care of this.
This could potentially clash with the raid6 library rework I'm doing
for 7.2. Although git has become pretty good about renamed files, so
maybe it won't be so bad.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox