From: Gopikrishna Garmidi <gopikrishna.garmidi@oss.qualcomm.com>
To: Konrad Dybcio <konradybcio@kernel.org>,
Bjorn Andersson <andersson@kernel.org>
Cc: linux-arm-msm@vger.kernel.org, linux-kernel@vger.kernel.org,
Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Subject: Re: [PATCH] soc: qcom: smem: support Glymur DRAM info
Date: Tue, 25 Aug 2026 15:54:47 +0530 [thread overview]
Message-ID: <cd434320-4bb4-4937-aa75-6f0dd081049a@oss.qualcomm.com> (raw)
In-Reply-To: <20260820-topic-glymur_dramc-v1-1-f714eef34e4d@oss.qualcomm.com>
On 8/20/2026 10:00 PM, Konrad Dybcio wrote:
> From: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
>
> Glymur firmware provides a version 7 DDR-info SMEM entry whose 872-byte
> layout is different from the one the driver supports today.
>
> The data contains 16 channel descriptors, 14 DDR frequency slots, four
> 4 DDR region slots, a 10-entry SHUB frequency plan, and v6 misc information.
> It accounts for 616 bytes; the captured SMEM entry has a further 256
> zero bytes.
>
> Add a Glymur-specific version 7 layout and recognize that complete entry.
> Parse the DDR frequencies and highest-bank bit from the common fields.
>
> Assisted-by: LLM
> Signed-off-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
> ---
Hi Konrad,
I tested this on the Glymur CRD.
The new v7 layout is parsed correctly: the "unknown type of DRAM info
struct (size = 872)" error is gone, and hbb reports 16.
The frequency list is almost correct, but the highest entry is truncated:
# cat /sys/kernel/debug/qcom_smem/dram_frequencies
200000000
547200000
1353600000
1555200000
1708800000
2092800000
2736000000
3187200000
3686400000
4224000000
466632704
The last value should be 4761600000. This is caused by a 32-bit overflow
in the frequency calculation:
dram->frequencies[dram->num_frequencies++] = 1000 * freq_khz;
Since freq_khz is a u32, the multiplication is performed as a 32-bit
operation before being assigned to the u64 array. Using 1000UL instead
fixes the overflow:
dram->frequencies[dram->num_frequencies++] = 1000UL * freq_khz;
The same issue exists in the v3, v4, v5, and v7 parsing functions. It
was not exposed previously because their maximum frequencies were below
the 32-bit limit.
With this change, the last frequency is reported correctly as
4761600000.
I can send the overflow fix as a separate patch, or you can fold it in
—whichever you prefer.
Tested-by: Gopikrishna Garmidi <gopikrishna.garmidi@oss.qualcomm.com>
Thanks,
Gopikrishna
next prev parent reply other threads:[~2026-08-25 10:24 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-20 16:30 Konrad Dybcio
2026-08-20 19:31 ` Abel Vesa
2026-08-24 14:38 ` Konrad Dybcio
2026-08-25 10:24 ` Gopikrishna Garmidi [this message]
2026-08-26 13:11 ` Konrad Dybcio
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=cd434320-4bb4-4937-aa75-6f0dd081049a@oss.qualcomm.com \
--to=gopikrishna.garmidi@oss.qualcomm.com \
--cc=andersson@kernel.org \
--cc=konrad.dybcio@oss.qualcomm.com \
--cc=konradybcio@kernel.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®