fs/pseudofile: fix buffer sizing and sparse growth

[YWHKKX]

Summary

• Fix incorrect in-memory buffer sizing when a pseudofile is extended.
• Grow the allocation by doubling instead of shift-based round-up.
• Clear newly addressed bytes on growth; truncate growth uses the same expand path.

Impact

• New/changed feature? NO (bug fix only).
• Impact on user / API? NO.
• Impact on build / hardware / documentation? NO.
• Impact on compatibility? NO.

Testing

I confirm that changes are verified on local setup and work as intended:

• Build Host: WSL2 Ubuntu 24.04 (x86_64), gcc 13 + gcc-multilib
• Target: arch=sim (32-bit) + KASAN, CONFIG_PSEUDOFS_FILE=y
• Repro: open a pseudofile, seek far ahead, then write

Before

open: OK
lseek: OK
write: Segmentation fault (rc=139)

After

open: OK
lseek: OK
write returned -1 errno=12

PR verification Self-Check

• This PR introduces only one functional change.
• I have updated all required description fields above.
• My PR adheres to Contributing Guidelines and Documentation.
• My PR is still work in progress (not ready for review).
• My PR is ready for review and can be safely merged into a codebase.

[github-actions]

MemBrowse Memory Report

hifive1-revb

• flash: .text +4 B (+0.0%, 83,428 B / 4,194,304 B, total: 2% used)
  No memory changes detected for:
• arduino-mega2560
• esp32-devkitc
• mirtoo
• qemu-armv8a
• qemu-intel64
• rx65n-rsk2mb
• s698pm-dkit
• stm32-nucleo-f103rb

[cederom]

Thank you @YWHKKX :-)

Please:

• sign the commit with git commit -s (see https://nuttx.apache.org/docs/latest/contributing/guide.html#commit-requirements).
• Update according to @xiaoxiang781216 hints.

[jerpelea]

please sign the patch
git commit -s

[xiaoxiang781216]

@YWHKKX please squash your change into one patch.

[YWHKKX]

@xiaoxiang781216 already squashed to c52c411

[linguini1]

Code review of the updated expand/write/truncate paths on current master.

Please provide some more testing detail. Did you perform any runtime tests?

[cederom]

Thank you @YWHKKX :-) Before and after runtime tests are welcome as @linguini1 noted :-)

[YWHKKX]

Thanks @linguini1 @cederom :-)

Yes — runtime tested on WSL2 with sim (32-bit) + KASAN, CONFIG_PSEUDOFS_FILE=y.

Before: write after a large seek crashes (SIGSEGV, rc=139).
After (this PR): write fails safely with -ENOMEM (errno=12); no crash.

[linguini1]

Can you please give a concrete test setup or runtime logs?

[YWHKKX]

Thanks @linguini1.

Concrete test setup

• Host: WSL2, Ubuntu 24.04, x86_64
• Toolchain: gcc 13 + gcc-multilib
• Repos: nuttx + nuttx-apps (side by side)
• Configure:
  ./tools/configure.sh -a ../nuttx-apps sim:
  (local sim defconfig enables at least:
  CONFIG_SIM_M32=y,
  CONFIG_MM_KASAN=y,
  CONFIG_MM_KASAN_INSTRUMENT_ALL=y,
  CONFIG_PSEUDOFS_FILE=y,
  and a small init-time test app)
• Build/run:
  $ make -j$(nproc)
  $ ./nuttx

The init-time test opens a pseudofile, seeks to a large offset,
then writes a small buffer.

Runtime logs

Before this PR:
open: OK
lseek: OK
write: Segmentation fault
exit status: 139

After this PR:
open: OK
lseek: OK
write returned -1 errno=12
(-ENOMEM, no crash)

[acassis]

@linguini1 PTAL, here already replied

Tests