From: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
To: catalin.marinas@arm.com, will@kernel.org
Cc: robin.murphy@arm.com, amit.pundir@linaro.org,
andersson@kernel.org, quic_sibis@quicinc.com,
sumit.semwal@linaro.org, linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] Revert "arm64: dma: Drop cache invalidation from arch_dma_prep_coherent()"
Date: Mon, 14 Nov 2022 16:59:35 +0530 [thread overview]
Message-ID: <20221114112935.GN3869@thinkpad> (raw)
In-Reply-To: <20221114110329.68413-1-manivannan.sadhasivam@linaro.org>
On Mon, Nov 14, 2022 at 04:33:29PM +0530, Manivannan Sadhasivam wrote:
> This reverts commit c44094eee32f32f175aadc0efcac449d99b1bbf7.
>
> As reported by Amit [1], dropping cache invalidation from
> arch_dma_prep_coherent() triggers a crash on the Qualcomm SM8250 platform
s/SM8250/SDM845/g
Sorry for the confusion.
Thanks,
Mani
> (most probably on other Qcom platforms too). The reason is, Qcom
> qcom_q6v5_mss driver copies the firmware metadata and shares it with modem
> for validation. The modem has a secure block (XPU) that will trigger a
> whole system crash if the shared memory is accessed by the CPU while modem
> is poking at it.
>
> To avoid this issue, the qcom_q6v5_mss driver allocates a chunk of memory
> with no kernel mapping, vmap's it, copies the firmware metadata and
> unvmap's it. Finally the address is then shared with modem for metadata
> validation [2].
>
> Now because of the removal of cache invalidation from
> arch_dma_prep_coherent(), there will be cache lines associated with this
> memory even after sharing with modem. So when the CPU accesses it, the XPU
> violation gets triggered.
>
> So let's revert this commit to get remoteproc's working (thereby avoiding
> full system crash) on Qcom platforms.
>
> [1] https://lore.kernel.org/linux-arm-kernel/CAMi1Hd1VBCFhf7+EXWHQWcGy4k=tcyLa7RGiFdprtRnegSG0Mw@mail.gmail.com/
> [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/remoteproc/qcom_q6v5_mss.c#n933
>
> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> ---
>
> Will, Catalin: Please share if you have any other suggestions to handle the
> resource sharing in the remoteproc driver that could avoid this revert.
>
> arch/arm64/mm/dma-mapping.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
> index 3cb101e8cb29..7d7e9a046305 100644
> --- a/arch/arm64/mm/dma-mapping.c
> +++ b/arch/arm64/mm/dma-mapping.c
> @@ -36,7 +36,7 @@ void arch_dma_prep_coherent(struct page *page, size_t size)
> {
> unsigned long start = (unsigned long)page_address(page);
>
> - dcache_clean_poc(start, start + size);
> + dcache_clean_inval_poc(start, start + size);
> }
>
> #ifdef CONFIG_IOMMU_DMA
> --
> 2.25.1
>
--
மணிவண்ணன் சதாசிவம்
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
WARNING: multiple messages have this Message-ID (diff)
From: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
To: catalin.marinas@arm.com, will@kernel.org
Cc: robin.murphy@arm.com, amit.pundir@linaro.org,
andersson@kernel.org, quic_sibis@quicinc.com,
sumit.semwal@linaro.org, linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] Revert "arm64: dma: Drop cache invalidation from arch_dma_prep_coherent()"
Date: Mon, 14 Nov 2022 16:59:35 +0530 [thread overview]
Message-ID: <20221114112935.GN3869@thinkpad> (raw)
In-Reply-To: <20221114110329.68413-1-manivannan.sadhasivam@linaro.org>
On Mon, Nov 14, 2022 at 04:33:29PM +0530, Manivannan Sadhasivam wrote:
> This reverts commit c44094eee32f32f175aadc0efcac449d99b1bbf7.
>
> As reported by Amit [1], dropping cache invalidation from
> arch_dma_prep_coherent() triggers a crash on the Qualcomm SM8250 platform
s/SM8250/SDM845/g
Sorry for the confusion.
Thanks,
Mani
> (most probably on other Qcom platforms too). The reason is, Qcom
> qcom_q6v5_mss driver copies the firmware metadata and shares it with modem
> for validation. The modem has a secure block (XPU) that will trigger a
> whole system crash if the shared memory is accessed by the CPU while modem
> is poking at it.
>
> To avoid this issue, the qcom_q6v5_mss driver allocates a chunk of memory
> with no kernel mapping, vmap's it, copies the firmware metadata and
> unvmap's it. Finally the address is then shared with modem for metadata
> validation [2].
>
> Now because of the removal of cache invalidation from
> arch_dma_prep_coherent(), there will be cache lines associated with this
> memory even after sharing with modem. So when the CPU accesses it, the XPU
> violation gets triggered.
>
> So let's revert this commit to get remoteproc's working (thereby avoiding
> full system crash) on Qcom platforms.
>
> [1] https://lore.kernel.org/linux-arm-kernel/CAMi1Hd1VBCFhf7+EXWHQWcGy4k=tcyLa7RGiFdprtRnegSG0Mw@mail.gmail.com/
> [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/remoteproc/qcom_q6v5_mss.c#n933
>
> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> ---
>
> Will, Catalin: Please share if you have any other suggestions to handle the
> resource sharing in the remoteproc driver that could avoid this revert.
>
> arch/arm64/mm/dma-mapping.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
> index 3cb101e8cb29..7d7e9a046305 100644
> --- a/arch/arm64/mm/dma-mapping.c
> +++ b/arch/arm64/mm/dma-mapping.c
> @@ -36,7 +36,7 @@ void arch_dma_prep_coherent(struct page *page, size_t size)
> {
> unsigned long start = (unsigned long)page_address(page);
>
> - dcache_clean_poc(start, start + size);
> + dcache_clean_inval_poc(start, start + size);
> }
>
> #ifdef CONFIG_IOMMU_DMA
> --
> 2.25.1
>
--
மணிவண்ணன் சதாசிவம்
next prev parent reply other threads:[~2022-11-14 11:31 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-14 11:03 [PATCH] Revert "arm64: dma: Drop cache invalidation from arch_dma_prep_coherent()" Manivannan Sadhasivam
2022-11-14 11:03 ` Manivannan Sadhasivam
2022-11-14 11:29 ` Manivannan Sadhasivam [this message]
2022-11-14 11:29 ` Manivannan Sadhasivam
2022-11-14 14:11 ` Will Deacon
2022-11-14 14:11 ` Will Deacon
2022-11-14 15:14 ` Robin Murphy
2022-11-14 15:14 ` Robin Murphy
2022-11-14 17:38 ` Catalin Marinas
2022-11-14 17:38 ` Catalin Marinas
2022-11-18 10:54 ` Manivannan Sadhasivam
2022-11-18 10:54 ` Manivannan Sadhasivam
2022-11-18 12:33 ` Will Deacon
2022-11-18 12:33 ` Will Deacon
2022-11-21 6:42 ` Manivannan Sadhasivam
2022-11-21 6:42 ` Manivannan Sadhasivam
2022-11-21 10:12 ` Sibi Sankar
2022-11-21 10:12 ` Sibi Sankar
2022-11-24 11:55 ` Catalin Marinas
2022-11-24 11:55 ` Catalin Marinas
2022-12-01 9:29 ` Thorsten Leemhuis
2022-12-01 9:29 ` Thorsten Leemhuis
2022-12-01 17:45 ` Catalin Marinas
2022-12-01 17:45 ` Catalin Marinas
2022-12-02 8:26 ` Amit Pundir
2022-12-02 8:26 ` Amit Pundir
2022-12-02 8:54 ` Thorsten Leemhuis
2022-12-02 8:54 ` Thorsten Leemhuis
2022-12-02 10:03 ` Will Deacon
2022-12-02 10:03 ` Will Deacon
2022-12-02 10:34 ` Thorsten Leemhuis
2022-12-02 10:34 ` Thorsten Leemhuis
2022-12-02 16:10 ` Greg KH
2022-12-02 16:10 ` Greg KH
2022-12-02 16:27 ` Thorsten Leemhuis
2022-12-02 16:27 ` Thorsten Leemhuis
2022-12-02 16:32 ` Greg KH
2022-12-02 16:32 ` Greg KH
2022-12-02 17:14 ` Manivannan Sadhasivam
2022-12-02 17:14 ` Manivannan Sadhasivam
2022-12-05 14:24 ` Will Deacon
2022-12-05 14:24 ` Will Deacon
2022-12-06 9:21 ` Manivannan Sadhasivam
2022-12-06 9:21 ` Manivannan Sadhasivam
2022-12-06 9:58 ` Will Deacon
2022-12-06 9:58 ` Will Deacon
2022-12-02 10:54 ` Manivannan Sadhasivam
2022-12-02 10:54 ` Manivannan Sadhasivam
2022-11-28 5:44 ` Thorsten Leemhuis
2022-11-28 5:44 ` Thorsten Leemhuis
2022-11-28 8:15 ` Manivannan Sadhasivam
2022-11-28 8:15 ` Manivannan Sadhasivam
2022-12-08 4:59 ` Leonard Lausen
2022-12-08 4:59 ` Leonard Lausen
-- strict thread matches above, loose matches on Subject: below --
2022-12-06 10:34 Will Deacon
2022-12-06 11:24 ` Manivannan Sadhasivam
2022-12-06 17:50 ` Catalin Marinas
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20221114112935.GN3869@thinkpad \
--to=manivannan.sadhasivam@linaro.org \
--cc=amit.pundir@linaro.org \
--cc=andersson@kernel.org \
--cc=catalin.marinas@arm.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=quic_sibis@quicinc.com \
--cc=robin.murphy@arm.com \
--cc=sumit.semwal@linaro.org \
--cc=will@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.