public inbox for igt-dev@lists.freedesktop.org
 help / color / mirror / Atom feed
From: "Joshi, Kunal1" <kunal1.joshi@intel.com>
To: "Murthy, Arun R" <arun.r.murthy@intel.com>,
	igt-dev@lists.freedesktop.org
Subject: Re: [i-g-t,1/6] lib/igt_edid: add EDID serial extraction helpers
Date: Mon, 9 Mar 2026 16:32:19 +0530	[thread overview]
Message-ID: <b9215c41-315c-4a1c-b091-3cef36b675b5@intel.com> (raw)
In-Reply-To: <64b4bad8-09ff-4c5c-81d8-7d6767f4c472@intel.com>

Hello Arun,

Thanks for looking at the patches,

On 09-03-2026 15:00, Murthy, Arun R wrote:
> On 26-02-2026 02:58, Kunal Joshi wrote:
>> Add helpers to extract serial identification from EDID data.
>>
>>   - edid_get_serial_string(): extract ASCII serial from descriptor 0xFF
>>   - edid_get_serial_number(): extract 32-bit serial from EDID header
>>   - edid_get_any_serial(): convenience helper that tries ASCII serial
>>     first, then falls back to 32-bit header serial formatted as hex
>>
>> These are needed by the connector helper and USB4 switch test suite
>> for EDID-based display verification after dock/undock events.
>>
>> Signed-off-by: Kunal Joshi <kunal1.joshi@intel.com>
>> ---
>>   lib/igt_edid.c | 91 ++++++++++++++++++++++++++++++++++++++++++++++++++
>>   lib/igt_edid.h |  5 +++
>>   2 files changed, 96 insertions(+)
>>
>> diff --git a/lib/igt_edid.c b/lib/igt_edid.c
>> index c68ccf671..e7fcbf0b9 100644
>> --- a/lib/igt_edid.c
>> +++ b/lib/igt_edid.c
>> @@ -249,6 +249,97 @@ void edid_get_monitor_name(const struct edid 
>> *edid, char *name, size_t name_size
>>       name[0] = '\0';
>>   }
>>   +/**
>> + * edid_get_serial_string:
>> + * @edid: EDID structure
>> + * @serial: buffer to write serial string
>> + * @serial_size: size of serial buffer; must be > 1
>> + *
>> + * Extract the serial string from EDID descriptor block type 0xFF.
>> + * The descriptor payload is a 13-byte fixed-width field; trailing
>> + * whitespace is stripped and the result is NUL-terminated.
>> + * @serial is left empty when no 0xFF descriptor is found.
>> + */
>> +void edid_get_serial_string(const struct edid *edid,
>> +                char *serial, size_t serial_size)
> We are trying to get the timings so rather than a generic name 
> serial_string, can this be replaced with get_detailed_timings ?
Spec section 3.10.3.1
defines tag FFh as Display Product Serial Number Descriptor, not a 
detailed timing descriptor

Section 3.10.3
Notes display descriptors are not a detailed timing.
>> +{
>> +    size_t copy_len, i, n;
>> +
>> +    igt_assert(edid);
>> +    igt_assert(serial);
>> +    igt_assert(serial_size > 1);
>> +    serial[0] = '\0';
>> +
>> +    for (i = 0; i < DETAILED_TIMINGS_LEN; i++) {
>> +        const uint8_t *d = (const uint8_t *)&edid->detailed_timings[i];
>> +
>> +        if (d[0] != 0x00 || d[1] != 0x00 || d[2] != 0x00)
>> +            continue;
>> +
>> +        if (d[3] != EDID_DETAIL_MONITOR_SERIAL || d[4] != 0x00)
>> +            continue;
>> +
>> +        /* Descriptor string payload is 13 bytes starting at d[5] */
>> +        copy_len = serial_size - 1 < 13 ? serial_size - 1 : 13;
>> +        memcpy(serial, &d[5], copy_len);
> Can strncpy() be used instead?
strncpy() is not a better fit here. The EDID serial field is a
fixed-width 13-byte payload, not a NUL-terminated C string. Copying the
bounded payload with memcpy(), then explicitly NUL-terminating the
destination and trimming trailing whitespace/newline, matches the EDID
encoding better.
>> +        serial[copy_len] = '\0';
>> +
>> +        for (n = strlen(serial);
>> +             n > 0 && isspace((unsigned char)serial[n - 1]); n--)
>> +            serial[n - 1] = '\0';
>> +
>> +        return;
>> +    }
>> +}
>> +
>> +/**
>> + * edid_get_serial_number:
>> + * @edid: EDID structure
>> + *
>> + * Extract the 32-bit serial number from the EDID base block header
>> + * (bytes 12–15, the manufacturer/product section).  This is different
>> + * from the ASCII serial string in descriptor block 0xFF.
>> + *
>> + * Returns: 32-bit serial number (little-endian), 0 if absent.
>> + */
>> +uint32_t edid_get_serial_number(const struct edid *edid)
>> +{
>> +    return (uint32_t)edid->serial[0] |
>> +           ((uint32_t)edid->serial[1] << 8) |
>> +           ((uint32_t)edid->serial[2] << 16) |
>> +           ((uint32_t)edid->serial[3] << 24);
>> +}
>> +
>> +/**
>> + * edid_get_any_serial:
> No description added?
Description of the function is present below

Thanks and Regards
Kunal Joshi
>
>
> Thanks and Regards,
> Arun R Murthy
> --------------------
>> + * @edid: EDID structure
>> + * @serial: buffer to write serial string
>> + * @serial_size: size of serial buffer; must be > 1
>> + *
>> + * Convenience helper that tries to extract a serial identifier from 
>> EDID.
>> + * First tries the ASCII serial string descriptor (0xFF). If that is 
>> empty,
>> + * falls back to the 32-bit header serial number formatted as 8-char 
>> hex.
>> + * If neither source provides a usable serial (no 0xFF descriptor 
>> and the
>> + * header serial is 0), @serial is left empty.
>> + */
>> +void edid_get_any_serial(const struct edid *edid,
>> +             char *serial, size_t serial_size)
>> +{
>> +    uint32_t serial_num;
>> +
>> +    igt_assert(edid);
>> +    igt_assert(serial);
>> +    igt_assert(serial_size > 1);
>> +
>> +    edid_get_serial_string(edid, serial, serial_size);
>> +    if (serial[0])
>> +        return;
>> +
>> +    serial_num = edid_get_serial_number(edid);
>> +    if (serial_num != 0)
>> +        snprintf(serial, serial_size, "%08X", serial_num);
>> +}
>> +
>>   static void edid_set_mfg(struct edid *edid, const char mfg[static 3])
>>   {
>>       edid->mfg_id[0] = (mfg[0] - '@') << 2 | (mfg[1] - '@') >> 3;
>> diff --git a/lib/igt_edid.h b/lib/igt_edid.h
>> index be0ccf529..d9e396065 100644
>> --- a/lib/igt_edid.h
>> +++ b/lib/igt_edid.h
>> @@ -457,5 +457,10 @@ void *dispid_block_tiled(void *ptr,
>>                int hsize, int vsize,
>>                const char *topology_id);
>>   void edid_get_monitor_name(const struct edid *edid, char *name, 
>> size_t name_size);
>> +void edid_get_serial_string(const struct edid *edid,
>> +                char *serial, size_t serial_size);
>> +uint32_t edid_get_serial_number(const struct edid *edid);
>> +void edid_get_any_serial(const struct edid *edid,
>> +             char *serial, size_t serial_size);
>>     #endif

  reply	other threads:[~2026-03-09 11:02 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-25 21:28 [PATCH i-g-t 0/6] add test to validate dock/undock and switch Kunal Joshi
2026-02-25 21:28 ` [PATCH i-g-t 1/6] lib/igt_edid: add EDID serial extraction helpers Kunal Joshi
2026-03-09  9:30   ` [i-g-t,1/6] " Murthy, Arun R
2026-03-09 11:02     ` Joshi, Kunal1 [this message]
2026-03-16  3:03       ` Murthy, Arun R
2026-02-25 21:28 ` [PATCH i-g-t 2/6] lib/igt_connector_helper: Add generic connector helpers Kunal Joshi
2026-03-09  9:38   ` [i-g-t,2/6] " Murthy, Arun R
2026-03-09 11:06     ` Joshi, Kunal1
2026-03-16  3:17       ` Murthy, Arun R
2026-02-25 21:28 ` [PATCH i-g-t 3/6] lib/igt_serial: add generic serial communication helper Kunal Joshi
2026-03-16  5:40   ` [i-g-t,3/6] " Murthy, Arun R
2026-02-25 21:28 ` [PATCH i-g-t 4/6] lib/igt_usb4_switch: add helper library for USB4 Switch 3141 Kunal Joshi
2026-03-16  8:45   ` [i-g-t,4/6] " Murthy, Arun R
2026-02-25 21:28 ` [PATCH i-g-t 5/6] tests/kms_feature_discovery: add basic usb4 switch discovery Kunal Joshi
2026-03-09 10:05   ` [i-g-t,5/6] " Murthy, Arun R
2026-02-25 21:28 ` [PATCH i-g-t 6/6] tests/intel/kms_usb4_switch: Add USB4 switch test suite Kunal Joshi
2026-03-09 10:03   ` [i-g-t,6/6] " Murthy, Arun R
2026-03-09 11:14     ` Joshi, Kunal1
2026-03-16  3:38       ` Murthy, Arun R
2026-02-26  2:20 ` ✗ Xe.CI.BAT: failure for add test to validate dock/undock and switch (rev3) Patchwork
2026-02-26  3:04 ` ✓ i915.CI.BAT: success " Patchwork
2026-02-26  6:52 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-02-26  7:45 ` ✗ i915.CI.Full: " Patchwork

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=b9215c41-315c-4a1c-b091-3cef36b675b5@intel.com \
    --to=kunal1.joshi@intel.com \
    --cc=arun.r.murthy@intel.com \
    --cc=igt-dev@lists.freedesktop.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