New kernels

[xeno74]

For this reason I only trust managing Amiga partitions from OS4 and only do formatting at the most from Linux when managing them.

Yes, this is correct way.

Partitons that did overflow the disk size (due to 32 bit int overflow) were not skipped but truncated to the end of the disk. Users who missed the warning message during boot would go on to create a filesystem with a size exceeding the actual partition size. Now that the 32 bit overflow has been corrected, such filesystems may refuse to mount with a 'filesystem exceeds partition size' error. Users should either correct the partition size, or resize the filesystem before attempting to boot a kernel with the RDB fixes in place.

Media Toolbox has created the RDB correctly but Linux doesn’t read it correctly.

[xeno74]

Hi All,

Here is the third alpha of kernel 6.5 for testing.

New:

• Linux 6.5 Brings Improvements To Firewire Support -- phoronix.com
• X1000 and X5000: FireWire support (CONFIG_FIREWIRE=y)
• USB / Thunderbolt driver updates
• NFS client updates
• Linux Git log
• Phoronix articles, reviews and news stories covering Linux 6.5

Download: linux-image-6.5-alpha3-X1000_X5000.tar.gz



Please test the kernels.

Thanks,
Christian

[xeno74]

FYI: The NTFS3 driver has also seen some CPU endian fixes.

Further information: NTFS Driver For Linux 6.5 Brings Small Optimizations & Fixes — phoronix.com

[xeno74]

FYI: The NTFS3 driver has also seen some CPU endian fixes.

Further information: NTFS Driver For Linux 6.5 Brings Small Optimizations & Fixes — phoronix.com

It seems, the issue is fixed. Issue: [Linux PPC] Issues with NTFS read-write driver (GPL implementation by Paragon Software)

[xeno74]

Hi All,

The merge window has been closed and I am happy to announce the RC1 of kernel 6.5 for testing.

New:

The issue with NTFS3 from Paragon Software is fixed You can now use NTFS3 to mount, read and write NTFS partitions.

More news:

• Linux 6.5-rc1 Released With Initial USB4 v2 Support, Cachestat, AMD RDNA3 Overclocking -- phoronix.com
• Release announcement
• NTFS Driver For Linux 6.5 Brings Small Optimizations & Fixes phoronix.com (including some CPU endian fixes)
• PowerPC fixes 6.5-2
• DMA mapping fixes
• XFS fix
• Sound fixes
• DRM fixes
• Linux Git log
• Phoronix articles, reviews and news stories covering Linux 6.5

Download: linux-image-6.5-rc1-X1000_X5000.tar.gz



Please test the kernels.

Thanks,
Christian

[Hypex]

Yes, this is correct way.

That's good. I must go through my logs as I didn't notice any RDSK errors.

Well I tested the alpha and all worked fine.

Unfortunately I upgraded my Fienix which ruined it but will open another thread about that.

[xeno74]

Well I tested the alpha and all worked fine.

Great! Thanks a lot for testing!

[xeno74]

Update amiga.c_v1.patch:

Code: Select all

--- a/block/partitions/amiga.c	2023-07-01 06:32:03.106124319 +0200
+++ b/block/partitions/amiga.c	2023-07-01 06:36:59.425257149 +0200
@@ -90,7 +90,7 @@ int amiga_partition(struct parsed_partit
 	}
 	blk = be32_to_cpu(rdb->rdb_PartitionList);
 	put_dev_sector(sect);
-	for (part = 1; blk>0 && part<=16; part++, put_dev_sector(sect)) {
+	for (part = 1; (s32) blk>0 && part<=16; part++, put_dev_sector(sect)) {	
 		/* Read in terms partition table understands */
 		if (check_mul_overflow(blk, (sector_t) blksize, &blk)) {
 			pr_err("Dev %s: overflow calculating partition block %llu! Skipping partitions %u and beyond\n",

I successfully created a RDB with GParted with the unpatched (reverted updates) RC1 of kernel 6.5 today.
After that, I tested this disk with a patched RC1 of kernel 6.5. (with Michael's patches)


The strange thing is, that there is no difference between a patched and an unpatched RC1. The reading of the RDB and partitions work correctly with both kernels.

Maybe my RDB was created incorrectly with Media Toolbox and the unpatched RC1 reports problems, but the partitons still work.

Screenshots:

[xeno74]

Hi All,

I was able to simulate the issue with the RDB.

First, I created 2TB dynamic allocation disk image with qemu-img.

Code: Select all

qemu-img create -f qcow2 disk.qcow2 2T

du -sh disk.qcow2

Code: Select all

352K    disk.qcow2

After that, I booted the virtual e5500 QEMU machine with this disk.

Code: Select all

qemu-system-ppc64 -M ppce500 -cpu e5500 -m 2048 -kernel uImage-6.5 -drive format=raw,file=void-live-powerpc-20230317.img,index=0,if=virtio -netdev user,id=mynet0 -device virtio-net,netdev=mynet0 -append "rw root=/dev/vda" -device virtio-gpu-pci -device virtio-mouse-pci -device virtio-keyboard-pci -device pci-ohci,id=newusb -audiodev id=sndbe,driver=pa,server=/run/user/1000/pulse/native -device usb-audio,bus=newusb.0 -smp 4 -fsdev local,security_model=passthrough,id=fsdev0,path=/home/christian/Dokumente -device virtio-9p-pci,id=fs0,fsdev=fsdev0,mount_tag=hostshare -drive format=qcow2,file=disk.qcow2,if=virtio

I was able to restore my RDB in this virtual e5500 QEMU machine with:

Code: Select all

sudo dd if=rdb-sda.img of=/dev/vdb

Code: Select all

sudo mkfs -t ext4 /dev/vdb4

I can mount this partition with the unpatched RC1. (without Michael's patches)
But I can't mount it with the patched RC1 of kernel 6.5. (with Michael's patches)

Code: Select all

... EXT4-fs (vdb4): bad geometry: block count ...

Michael's fixed RDB for my X1000: rdb-sda-edit

[Hypex]

I've just noticed I cannot see any instructions to instal the kernels. I'm sure this used to be in kernel threads but the latest ones are missing. Is there are guide around?

I was checking out my modules and found I could simply delete some. Which doesn't seem right. Usually it gives me a permission error.