Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-12 Thread Thomas Schmitt
Hi, Anatoly Pugachev wrote: > mator@chubaka:~$ md5sum /tmp/half_sparse /mnt/iso/half_sparse > 3e98c342558c453b379cf7699f98a108 /tmp/half_sparse > Segmentation fault Hah ! The obvious bug about all-zero blocks has bitten. I would not have expected a memory fault, though. Rather some false output

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-12 Thread Anatoly Pugachev
On Wed, Nov 11, 2020 at 11:09 PM Thomas Schmitt wrote: > > Hi, > > after some confusion i found out that the endianness bug was fixed > unnoticed a few weeks ago, when i implemented zisofs version 2. The > current GNU xorriso development tarball has it already fixed. >

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-11 Thread Thomas Schmitt
Hi, after some confusion i found out that the endianness bug was fixed unnoticed a few weeks ago, when i implemented zisofs version 2. The current GNU xorriso development tarball has it already fixed. https://www.gnu.org/software/xorriso/xorriso-1.5.3.tar.gz Anatoly Pugachev wrote: > copied

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-11 Thread Anatoly Pugachev
On Wed, Nov 11, 2020 at 8:13 PM Thomas Schmitt wrote: > > Hi, > > Anatoly Pugachev wrote directly to me: > > sent you an ISO file with another account... > > Now that's embrassing. My zisofs compressor did put out the zisofs > block pointers into the file as big-endian 32-bit numbers, where >

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-11 Thread Thomas Schmitt
Hi, Anatoly Pugachev wrote directly to me: > sent you an ISO file with another account... Now that's embrassing. My zisofs compressor did put out the zisofs block pointers into the file as big-endian 32-bit numbers, where everybody expects little-endian. So the bread() call in the kernel which

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-11 Thread Thomas Schmitt
Hi, Anatoly Pugachev wrote: > copied to my local x86 VM: > mator@u164:~> sudo mount -o loop,ro test.iso /mnt > mator@u164:~> md5sum /mnt/1mb > md5sum: /mnt/1mb: Input/output error Grrr. This is not supposed happen. (I tested this here. Grrr, again.) > probably xorriso wrote a wrong iso on ppc64

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-11 Thread Anatoly Pugachev
On Wed, Nov 11, 2020 at 3:26 PM Thomas Schmitt wrote: > > Hi, > > Anatoly Pugachev wrote: > > mator@chubaka:~$ getconf PAGESIZE > > 65536 > > mator@chubaka:~$ uname -a > > Linux chubaka 5.9.0-1-powerpc64 #1 SMP Debian 5.9.1-1 (2020-10-17) > > ppc64 GNU/Linux > > [...] > > mator@chubaka:~$ md5sum

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-11 Thread Thomas Schmitt
Hi, to answer my own question whether bytes were delivered: The strace shows that there were none. The first read of 32 KiB already fails. Hopefully a useful hint. Have a nice day :) Thomas

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-11 Thread Thomas Schmitt
Hi, Anatoly Pugachev wrote: > mator@chubaka:~$ getconf PAGESIZE > 65536 > mator@chubaka:~$ uname -a > Linux chubaka 5.9.0-1-powerpc64 #1 SMP Debian 5.9.1-1 (2020-10-17) > ppc64 GNU/Linux > [...] > mator@chubaka:~$ md5sum /mnt/1mb > md5sum: /mnt/1mb: Input/output error Does it deliver any bytes

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-11 Thread Anatoly Pugachev
On Wed, Nov 11, 2020 at 11:32 AM Thomas Schmitt wrote: > Anatoly Pugachev wrote: > > > > $ ssh gcc203 getconf PAGESIZE > > > > 65536 > I wrote: > > > If there is a machine of that kind which can be used for a short bitrot > > > test of the existing kernel with CONFIG_ZISOFS=y (which is usually

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-11 Thread Thomas Schmitt
Hi, Anatoly Pugachev wrote: > > > $ ssh gcc203 getconf PAGESIZE > > > 65536 I wrote: > > If there is a machine of that kind which can be used for a short bitrot > > test of the existing kernel with CONFIG_ZISOFS=y (which is usually already > > set), i'd be glad to propose a few shell commands

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-10 Thread Anatoly Pugachev
On Tue, Nov 10, 2020 at 9:20 PM Thomas Schmitt wrote: > Anatoly Pugachev wrote: > > $ ssh gcc203 getconf PAGESIZE > > 65536 > > [...] > > thouse gcc* machines are from 'gcc compile farm' [1] , ppc64 or ppc64le > > Ah yes. That's probably what i ask about. > arch/powerpc/include/asm/page.h >

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-10 Thread Thomas Schmitt
Hi, Anatoly Pugachev wrote: > $ ssh gcc203 getconf PAGESIZE > 65536 > [...] > thouse gcc* machines are from 'gcc compile farm' [1] , ppc64 or ppc64le Ah yes. That's probably what i ask about. arch/powerpc/include/asm/page.h /* ... For PPC64 we support either 4K or 64K software page size. ...

Re: Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-10 Thread Anatoly Pugachev
On Mon, Nov 9, 2020 at 9:41 PM Thomas Schmitt wrote: > > Hi, > > by incident one of my other kernel adventures leads me here, although > it might turn out to be off topic, as i know too few about powerpc. > > Are there machines in use which really have regular memory page size > larger than 32

Are there still machines with PAGE_SHIFT 16 or larger ?

2020-11-09 Thread Thomas Schmitt
Hi, by incident one of my other kernel adventures leads me here, although it might turn out to be off topic, as i know too few about powerpc. Are there machines in use which really have regular memory page size larger than 32 KiB ? I read in Linux source arch/powerpc/include/asm/page.h this