[dimerr]

Same change as #171, for hi3519dv500.

Change

Quad gives more problems than benefits on these cameras. The OpenIPC config leaves CLOSE_SPI_8PIN_4IO unset, so bsp-sfc offers quad here. This removes the quad modes at probe and at resume. spi_nor_quad_enable() then returns early, so SR is not written for it.

CLOSE_SPI_8PIN_4IO no longer has any effect on this branch.

Testing

Build-tested only: drivers/mtd/spi-nor with the hi3519dv500 config from OpenIPC/firmware. Please check on hardware:

• A board boots and a whole-partition read gives a stable md5sum without quad.
• A Winbond or GigaDevice board still boots. With no quad in hwcaps, spi_nor_dtrclear_pinreset() now reaches spi_nor_sr3_to_reset() there. It writes SR3 with no WREN first, so it should do nothing with WEL clear.

[qodo-free-for-open-source-projects]

PR Summary by Qodo

hi3519dv500: Disable BSP SFC quad modes at probe and resume

🐞 Bug fix 🕐 10-20 Minutes

AI Description

• Remove quad read and program capabilities from BSP SFC probe and resume to avoid unreliable camera
 operation.
• Keep quad-enable status-register writes from running when a non-quad protocol is selected.
• Hardware validation remains needed, particularly for Winbond and GigaDevice reset-pin handling.

Diagram

graph TD
  Probe["Device probe"] --> Caps["Non-quad capabilities"] --> Setup["Protocol setup"] --> Protocol["Single/dual protocol"] --> Quad["Quad enable skipped"]
  Resume["Device resume"] --> Caps
  Caps --> Reset["Reset-pin handling"]

Loading

High-Level Assessment

Removing quad from the controller capabilities at both entry points is the direct way to prevent its selection regardless of CLOSE_SPI_8PIN_4IO. Changing flash-specific parameters instead would not express this controller-level restriction as clearly. Validate whole-partition reads and Winbond/GigaDevice boot behavior on hardware before merging.

Files changed (2) +0 / -14

Bug fix (2) +0 / -14

bsp-sfc.cStop advertising quad modes during BSP SFC probe +0/-7

Stop advertising quad modes during BSP SFC probe

• Removes the conditional addition of quad read and page-program capabilities before SPI NOR scanning. The configuration option no longer controls quad capability advertisement at this entry point.

drivers/mtd/spi-nor/controllers/bsp-sfc.c

core.cKeep quad modes disabled during BSP SFC resume +0/-7

Keep quad modes disabled during BSP SFC resume

• Removes the corresponding conditional quad capabilities from the resume path, so protocol setup does not reselect quad after suspend. Without quad read capabilities, the existing manufacturer-specific reset-pin path can run for Winbond and GigaDevice parts.

drivers/mtd/spi-nor/core.c

[qodo-free-for-open-source-projects]

Code Review by Qodo

🐞 Bugs (1) 📘 Rule violations (0) 📎 Requirement gaps (0) 🎨 UX issues (0) 🔗 Cross-repo conflicts (0) 📜 Skill insights (0)

1. Reset pins stay disabled on some flashes 🐞 Bug ≡ Correctness

Description

spi_nor_sr3_to_reset() sends the status-register write without first enabling writes, then checks
the preceding read result instead of the write result. When a Winbond or GigaDevice flash has its
reset-enable bit clear and its write-enable latch clear, the newly reached path at probe and resume
reports success while leaving that bit unchanged.

Code

drivers/mtd/spi-nor/controllers/bsp-sfc.c[L589-592]

-	hwcaps.mask |= SNOR_HWCAPS_READ_1_1_4 |
-		       SNOR_HWCAPS_READ_1_4_4 |
-		       SNOR_HWCAPS_PP_1_1_4 |
-		       SNOR_HWCAPS_PP_1_4_4;

Evidence

The removed probe capabilities and matching resume mask cause the no-quad condition to enter the
reset helper. That helper dispatches Winbond and GigaDevice flashes to an SR3 writer that neither
sends write-enable nor captures the write return; the existing status-register write routine
demonstrates the required write-enable sequence.

drivers/mtd/spi-nor/controllers/bsp-sfc.c[527-542]
drivers/mtd/spi-nor/core.c[3410-3444]
drivers/mtd/spi-nor/core.c[2369-2405]
drivers/mtd/spi-nor/core.c[2456-2463]
drivers/mtd/spi-nor/core.c[724-740]
drivers/mtd/spi-nor/core.c[3644-3649]

Agent prompt

The issue below was found during a code review. Follow the provided context and guidance below and implement a solution

## Issue description
Removing quad capabilities makes probe and resume enter the reset-pin helper on Winbond and GigaDevice flashes. Its status-register write lacks write-enable and its result is discarded, so it can report success without enabling the reset pin.
## Fix Focus Areas
- drivers/mtd/spi-nor/core.c[2369-2395]
- drivers/mtd/spi-nor/core.c[2456-2463]
- drivers/mtd/spi-nor/core.c[3439-3444]
- drivers/mtd/spi-nor/core.c[3644-3649]
## Recommended Fix
If enabling RESET# is required, issue write-enable before the SR3 write, propagate the write result, wait for completion, and verify the bit. If this change must not touch SR3, avoid entering the reset-pin helper rather than treating an ineffective write as success.

ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools

---

Tip of the day

💡 Did you know, you can switch off images and animations for a plain-text comment

More tips ↗ | Customize Qodo ↗ | Qodo docs ↗