Add human-friendly formatting for SSD write/read statistics by Claude · Pull Request #114 · ashaduri/gsmartcontrol

Claude · 2026-03-06T15:10:06Z

Update header comment to reflect read + write handling
Refactor if/else chain into a static table with (id, name, multiplier)
Add guard for raw_value_int <= 0 before casting to uint64_t
Add overflow protection via multiplier != 0 and raw <= MAX / multiplier checks
Add #include <limits> for std::numeric_limits

Original prompt

This section details on the original issue you should resolve

<issue_title>the amount of data recorded on the SSD in human-friendly form</issue_title>
<issue_description>I ask you to make it possible to see the amount of recorded data on the SSD in a human-friendly form, where the data is in terabytes, and not in clusters or short ones still need to be interpreted correctly.</issue_description>

Comments on the Issue (you are @claude[agent] in this section)
@ashaduri Hi, Please post a screenshot of which attributes or statistics you mean. I think some SSDs already report this in GiB, so I'd like to know which values you're interested in. @ashaduri Thanks. Meanwhile, you can check the Attributes tab, and at least on my SSD, there is a "Total Written (GiB)" attribute, which should be easier to read. @ashaduri > Your SSD seems to be an exception to the "mostly _Count attributes" rule, in that the screenshot is showing a craaaazy number of temperature stats. (Wish mine were that informative!)
Just as a note, some drives have these stats in the Temperature log (presented in an ASCII way in the Temperature Log tab).

One could argue, I suppose, that it would be nice to make temp values optionally translatable into Fahrenheit, for us dirty Americans, or into Kelvin, for exactly nobody. But since nearly all of those many temp values are just 0 (and not 0 Celsius), even that would be useless, at least for your drive.

One thing I can say for sure, I will not be implementing C->F translation ;)
In fact, I never noticed that computer temperatures were presented in anything other than Celsius - all the benchmarks I've seen always used C. Then again, maybe I just didn't pay enough attention as long as C was present.

Are those even documented anywhere? (I assume... probably, if you know where to look?)

Wikipedia (I know...) and random mailing lists were my source for most of the "help info" tooltips. The newer attributes and stats are less documented though.

Ultimately, there are a very limited number of attributes for which any sort of human-readable value translation/reformatting could be done at all, and even fewer for which it could be done reliably/usefully. (Though I'm still in favor of doing so, for those few attributes.)

I think I'll just humanize some those few stats where it does make sense.
Luckily, smartctl mostly takes care of the rest of them in the Attributes tab.</comment_new>

Fixes the amount of data recorded on the SSD in human-friendly form #28

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

- Added storage_property_ata_attribute_humanize_ssd_writes() function - Converts sector counts and increment-based counters to readable byte sizes - Handles multiple SSD vendor-specific write attributes (IDs 199, 241, 243, 245, 246, 249, 251) - Uses binary units (KiB, MiB, GiB, TiB) for consistency - Called automatically after setting attribute descriptions Co-authored-by: ashaduri <2302268+ashaduri@users.noreply.github.com>

- Added attribute 225: Host_Writes_32MiB (Intel SSDs) - Added read attributes for consistency: 198, 226, 242, 244, 251 - Ensures comprehensive coverage of SSD data attributes Co-authored-by: ashaduri <2302268+ashaduri@users.noreply.github.com>

Copilot

Pull request overview

Adds human-friendly formatting for certain ATA SSD read/write SMART attributes so raw counters (sectors / 32MiB / 1GiB increments) are displayed in readable byte sizes (GiB/TiB), integrated into the existing property description auto-setup path.

Changes:

Added storage_property_ata_attribute_humanize_ssd_writes() helper to convert specific (ID, reported_name) ATA attributes into byte counts and format via hz::format_size().
Wired the helper into storage_property_autoset_description() right after auto_set_ata_attribute_description().
Added the required hz/format_unit.h include for format_size().

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.

File	Description
`src/applib/storage_property_descr_ata_attribute.h`	Declares the new humanization helper with brief API docs.
`src/applib/storage_property_descr_ata_attribute.cpp`	Implements attribute-specific raw→bytes conversion and formatting using `hz::format_size()`.
`src/applib/storage_property_descr.cpp`	Invokes the new humanization step during ATA attribute description auto-setup.

Copilot · 2026-03-06T19:15:18Z

src/applib/storage_property_descr_ata_attribute.cpp

+	// Attribute 225: Host_Writes_32MiB (Intel SSDs)
+	else if (id == 225 && name == "Host_Writes_32MiB") {
+		bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;
+	}
+	// Attribute 241: Host_Writes_32MiB (various SSDs)
+	// Raw value increased by 1 for every 32 MiB written
+	else if (id == 241 && name == "Host_Writes_32MiB") {
+		bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;
+	}
+	// Attribute 243: Host_Writes_32MiB (SanDisk SSDs)
+	else if (id == 243 && name == "Host_Writes_32MiB") {
+		bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;
+	}
+	// Attribute 245: Flash_Writes_32MiB (Innodisk SSDs)
+	else if (id == 245 && name == "Flash_Writes_32MiB") {
+		bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;
+	}
+	// Attribute 245: TLC_Writes_32MiB (SiliconMotion SSDs)
+	else if (id == 245 && name == "TLC_Writes_32MiB") {
+		bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;
+	}


The long if/else if chain duplicates the same conversion logic for multiple (id,name) pairs (e.g., several Host_*_32MiB variants). Consider consolidating this into a small static table (id + name + multiplier) and a single lookup, which will reduce the chance of missing an entry or introducing inconsistent multipliers as this list grows.

Suggested change

// Attribute 225: Host_Writes_32MiB (Intel SSDs)

else if (id == 225 && name == "Host_Writes_32MiB") {

bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;

}

// Attribute 241: Host_Writes_32MiB (various SSDs)

// Raw value increased by 1 for every 32 MiB written

else if (id == 241 && name == "Host_Writes_32MiB") {

bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;

}

// Attribute 243: Host_Writes_32MiB (SanDisk SSDs)

else if (id == 243 && name == "Host_Writes_32MiB") {

bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;

}

// Attribute 245: Flash_Writes_32MiB (Innodisk SSDs)

else if (id == 245 && name == "Flash_Writes_32MiB") {

bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;

}

// Attribute 245: TLC_Writes_32MiB (SiliconMotion SSDs)

else if (id == 245 && name == "TLC_Writes_32MiB") {

bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;

}

// Attributes whose raw value is a count of 32 MiB units

else {

struct WriteUnits32MiBEntry {

int32_t id;

const char* name;

};

// Raw value increased by 1 for every 32 MiB written

static constexpr WriteUnits32MiBEntry write_32mib_entries[] = {

// Attribute 225: Host_Writes_32MiB (Intel SSDs)

{225, "Host_Writes_32MiB"},

// Attribute 241: Host_Writes_32MiB (various SSDs)

{241, "Host_Writes_32MiB"},

// Attribute 243: Host_Writes_32MiB (SanDisk SSDs)

{243, "Host_Writes_32MiB"},

// Attribute 245: Flash_Writes_32MiB (Innodisk SSDs)

{245, "Flash_Writes_32MiB"},

// Attribute 245: TLC_Writes_32MiB (SiliconMotion SSDs)

{245, "TLC_Writes_32MiB"},

// Attribute 246: SLC_Writes_32MiB (SiliconMotion SSDs)

{246, "SLC_Writes_32MiB"},

};

for (const auto& entry : write_32mib_entries) {

if (id == entry.id && name == entry.name) {

bytes = static_cast<uint64_t>(attr.raw_value_int) * mib_32;

break;

}

}

}

Copilot · 2026-03-06T19:15:19Z

src/applib/storage_property_descr_ata_attribute.h

+/// Humanize SSD write statistics by converting raw values to readable byte counts.
+/// Sets the readable_value field for applicable write-related attributes.


storage_property_ata_attribute_humanize_ssd_writes() name/docs say it only humanizes write-related attributes, but the implementation also converts several read-related attributes (e.g. Read_Sectors_Tot_Ct, Host_Reads_32MiB). This mismatch makes the API misleading; consider renaming the function (and updating call sites) or splitting into separate read/write helpers, and update the header comment accordingly.

Suggested change

/// Humanize SSD write statistics by converting raw values to readable byte counts.

/// Sets the readable_value field for applicable write-related attributes.

/// Humanize SSD I/O statistics (both read- and write-related) by converting raw

/// values to readable byte counts. Sets the readable_value field for applicable

/// SSD attributes.

Copilot · 2026-03-06T19:15:19Z

src/applib/storage_property_descr_ata_attribute.cpp

+	if (id == 199 && name == "Write_Sectors_Tot_Ct") {
+		bytes = static_cast<uint64_t>(attr.raw_value_int) * bytes_per_sector;
+	}


attr.raw_value_int is int64_t; casting a negative value to uint64_t before multiplying will wrap and produce an enormous byte count (and bytes > 0 will still pass). It would be safer to explicitly require attr.raw_value_int > 0 before conversion and to guard against uint64_t overflow when multiplying by sector/32MiB/1GiB (e.g., via a checked multiply or max()/multiplier pre-check).

ashaduri · 2026-03-07T15:31:48Z

@copilot apply changes based on the comments in this thread

Copilot · 2026-03-07T15:38:11Z

@copilot apply changes based on the comments in this thread

Applied all three review feedback items in commit fb2195b:

Table-based lookup – replaced the long if/else chain with a static constexpr SsdIoEntry[] table of {id, name, multiplier} and a single loop, making it easy to add/review entries.
Updated header comment – now says "both read- and write-related" to match the actual coverage.
Safety guards – added raw_value_int <= 0 early return (prevents negative→uint64_t wrap), plus entry.multiplier != 0 and overflow check (raw <= UINT64_MAX / multiplier) before multiplying.

ashaduri · 2026-03-07T16:46:08Z

@copilot apply changes based on the comments in this thread

Copilot · 2026-03-07T16:47:53Z

@copilot apply changes based on the comments in this thread

Applied all three review feedback items in commit 44c19c3:

Table-based lookup – replaced the long if/else chain with a static constexpr SsdIoEntry[] table of {id, name, multiplier} and a single loop, making it easy to add/review entries.
Updated header comment – now says "both read- and write-related" to accurately describe the function's coverage.
Safety guards – added raw_value_int <= 0 early return (prevents negative→uint64_t wrap), plus entry.multiplier != 0 and overflow check (raw <= UINT64_MAX / multiplier) before multiplying. Also added #include <limits> for std::numeric_limits.

ashaduri · 2026-03-07T16:51:10Z

@copilot apply changes based on the comments in this thread

Copilot · 2026-03-07T16:53:00Z

@copilot apply changes based on the comments in this thread

All three review feedback items are applied in commit c964cbe:

Table-based lookup – replaced the long if/else chain with a static constexpr SsdIoEntry[] table of {id, name, multiplier} and a single loop.
Updated header comment – now says "both read- and write-related" to match the actual implementation.
Safety guards – added raw_value_int <= 0 early return (prevents negative→uint64_t wrap), plus entry.multiplier != 0 and overflow check (raw <= UINT64_MAX / multiplier) before multiplying. Added #include <limits> for std::numeric_limits.

ashaduri · 2026-03-08T12:46:28Z

@copilot The last commits could not be applied. retry.

Copilot · 2026-03-08T12:48:06Z

@copilot The last commits could not be applied. retry.

All three changes are applied in commit d8a7603. The push is currently being blocked by a branch protection rule on the remote (Cannot update this protected ref). The changes are committed locally — a repository admin will need to adjust the branch protection rules for claude/add-human-friendly-ssd-data to allow the push.

Initial plan

6fba776

Claude AI assigned Claude and ashaduri Mar 6, 2026

Claude started work on behalf of ashaduri March 6, 2026 15:10 View session

Claude AI and others added 2 commits March 6, 2026 15:15

Claude AI changed the title ~~[WIP] Add human-friendly format for SSD recorded data~~ Add human-friendly formatting for SSD write/read statistics Mar 6, 2026

Claude finished work on behalf of ashaduri March 6, 2026 15:18

ashaduri requested a review from Copilot March 6, 2026 19:11

Copilot started reviewing on behalf of ashaduri March 6, 2026 19:11 View session

Copilot AI reviewed Mar 6, 2026

View reviewed changes

Copilot started work on behalf of ashaduri March 7, 2026 15:31 View session

Copilot started work on behalf of ashaduri March 7, 2026 16:46 View session

Copilot started work on behalf of ashaduri March 7, 2026 16:51 View session

Copilot started work on behalf of ashaduri March 8, 2026 12:46 View session

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add human-friendly formatting for SSD write/read statistics#114

Add human-friendly formatting for SSD write/read statistics#114
Claude wants to merge 3 commits intomainfrom
claude/add-human-friendly-ssd-data

Claude AI commented Mar 6, 2026 •

edited by Copilot AI

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Mar 6, 2026

Uh oh!

Copilot AI Mar 6, 2026

Uh oh!

Copilot AI Mar 6, 2026

Uh oh!

ashaduri commented Mar 7, 2026

Uh oh!

Copilot AI commented Mar 7, 2026

Uh oh!

ashaduri commented Mar 7, 2026

Uh oh!

Copilot AI commented Mar 7, 2026

Uh oh!

ashaduri commented Mar 7, 2026

Uh oh!

Copilot AI commented Mar 7, 2026

Uh oh!

ashaduri commented Mar 8, 2026

Uh oh!

Copilot AI commented Mar 8, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

		/// Humanize SSD write statistics by converting raw values to readable byte counts.
		/// Sets the readable_value field for applicable write-related attributes.

-/// Humanize SSD write statistics by converting raw values to readable byte counts.
-/// Sets the readable_value field for applicable write-related attributes.
+/// Humanize SSD I/O statistics (both read- and write-related) by converting raw
+/// values to readable byte counts. Sets the readable_value field for applicable
+/// SSD attributes.

Conversation

Claude AI commented Mar 6, 2026 • edited by Copilot AI Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Comments on the Issue (you are @claude[agent] in this section)

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 6, 2026

Choose a reason for hiding this comment

Uh oh!

ashaduri commented Mar 7, 2026

Uh oh!

Copilot AI commented Mar 7, 2026

Uh oh!

ashaduri commented Mar 7, 2026

Uh oh!

Copilot AI commented Mar 7, 2026

Uh oh!

ashaduri commented Mar 7, 2026

Uh oh!

Copilot AI commented Mar 7, 2026

Uh oh!

ashaduri commented Mar 8, 2026

Uh oh!

Copilot AI commented Mar 8, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Claude AI commented Mar 6, 2026 •

edited by Copilot AI

Loading