Board index Linux development

Kernel change summary 1.1.92 -> 1.1.93

Kernel change summary 1.1.92 -> 1.1.93

Post by Russell Nels » Wed, 15 Mar 1995 11:37:38



These changes are archived on ftp.emlist.com:pub/kchanges and
http://www.nvg.unit.no/linux-changes/index.html.  I cannot answer
questions about Linux -- I merely summarize the kernel patches after
reading them.  I try to make them useful to everyone, but kernel
hackers should take them with a grain of salt and read the patches
themselves.  I'm not going to bother summarizing the Sparc, m68k, and
Alpha architecture changes until the respective port is working.

Check for a BIOS32 (PCI) Service Directory entry that's not in the low 1MB.
Changes to the Sparc entry point code.
Changes to the Sparc IRQ handler.
IDE README says that "single" drives should be jumpered as "master", not "slave".
New IDE driver, Version 3.12, simplifies logic for selecting initial
        mult_count (fixes problems with buggy WD drives)
Whoops!  We weren't doing the 250ms selection timeout properly for the
        53c7,8xx SCSI driver.
PCMCIA code for QLogic improved to version 0.40a.
Added "HITACHI","DK312C","CM81" to SCSI blacklist (responds to all LUN).
Use kmalloc for SCSI structures instead of local variables, but keep
        the code around in case it's needed (under
        USE_STATIC_SCSI_MEMORY).
Reduce SCSI disk latency by unconditionally turning interrupts on.
Recognize SCSI CD drives that don't do multisession.
Print an error on any SCSI tape abnormal conditions.
Differentiate between SCSI tape read retries and write retries
        (default to 0 write retries).
Be sure to free kernel memory onto the right free memory list.
Fix a bug in the memory mapper code.
Clean up code in swapper.
Do more work on closing sockets reliably.
Some Emacs variables (see below) added to SCSI files so that the
        tabbing would stay consistent with the One True Master's style.

/*
 * Overrides for Emacs so that we follow Linus's tabbing style.
 * Emacs will notice this stuff at the end of the file and automatically
 * adjust the settings for this buffer only.  This must remain at the end
 * of the file.
 * ---------------------------------------------------------------------------
 * Local variables:
 * c-indent-level: 8
 * c-brace-imaginary-offset: 0
 * c-brace-offset: -8
 * c-argdecl-indent: 8
 * c-label-offset: -8
 * c-continued-statement-offset: 8
 * c-continued-brace-offset: 0
 * End:
 */
--

Crynwr Software   | Crynwr Software sells packet driver support | ask4 PGP key
11 Grant St.      | +1 315 268 1925 (9201 FAX)  | What is thee doing about it?
Potsdam, NY 13676 | Capitalists try to avoid a free market.  Why might that be?

 
 
 
Top

Kernel change summary 1.1.92 -> 1.1.93

Post by da.. » Wed, 15 Mar 1995 13:32:53


 : IDE README says that "single" drives should be jumpered as "master", not "slave".
 : New IDE driver, Version 3.12, simplifies logic for selecting initial
 :      mult_count (fixes problems with buggy WD drives)

Which WD drives are buggy?  How do the bugs manifest themselves?

Thanks,
--John

 
 
 
Top

Kernel change summary 1.1.92 -> 1.1.93

Post by Mark Lo » Wed, 15 Mar 1995 23:36:40




> : IDE README says that "single" drives should be jumpered as "master", not "slave".
> : New IDE driver, Version 3.12, simplifies logic for selecting initial
> :  mult_count (fixes problems with buggy WD drives)

>Which WD drives are buggy?

Virtually all of their current models.  

Quote:>How do the bugs manifest themselves?

Numerous drive status errors during kernel initialization.

The problem is that the drives falsely report a "valid" multiplemode
setting, when in fact, their internal state has multiplemode disabled.

The fix is that the kernel now ignores what the drive reports,
and forces multiplemode off (by default) during kernel initialization.
--

For the latest Linux kernel: ftp.cs.helsinki.fi:/pub/Software/Linux/Kernel/v1.2
For Linux IDE (big/many) help, see:  linux-1.2.0/drivers/block/README.ide

 
 
 
Top

1. Kernel change summary 1.1.93 -> 1.1.94

These changes are archived on ftp.emlist.com:pub/kchanges and
http://www.nvg.unit.no/linux-changes/index.html.  I cannot answer
questions about Linux -- I merely summarize the kernel patches after
reading them.  I try to make them useful to everyone, but kernel
hackers should take them with a grain of salt and read the patches
themselves.  I'm not going to bother summarizing the Sparc, m68k, and
Alpha architecture changes until the respective port is working.

More Alpha changes.
i386 threading changed to not clone.  Instead a different syscall is
        used to clone.
Ptrace, shared memory, and proc fs code added to look for the page
        middle.
Sony CDU-535 driver cleaned up.
Wavelan driver cleaned up, processor-independent delays added.
When shrinking filesystem buffers, we didn't look at enough buffers.
Add a new mount option for ISO9660 fs -- "unhide" hidden files.
Removed unnecessary free list handling in lock handler.
Added unsigned long casts to ntohl() and ntohs() (portability).
Swapping code changed.
Stop destroying dead sockets when we bind a new one.
Don't initialize the TCP retransmit timer twice.

Serial handler changes:
rely on the line dispcline to prevent echo wars.
IEXTEN does not need to be enabled in order for IXANY to be active.
Change the interrupt poll time from 60 seconds to 10 seconds,
        configurable with the #define RS_STROBE_TIME.
Add provisions for a new type of interrupt service routine, which
        better supports multiple serial ports on a single IRQ.  
Add support to detect the StarTech 16650 chip.  Treat it as a 16450
        for now, because of its FIFO bugs.
Support close_wait in the serial driver (see below).

Close wait:
This is helpful for slow devices (like serial
plotters) so that their outputs don't get flushed upon
device close.  This has to be configurable because
normally we don't want ports to be hung up for long
periods of time during a close when they are not
connected to a device, or the device is powered off.

The default is to wait 30 seconds; in the case of a
very slow device, the close_wait timeout should be
lengthed.  If it is set to 0, the kernel will wait
forever for all of the data to be transmitted.

--

Crynwr Software   | Crynwr Software sells packet driver support | ask4 PGP key
11 Grant St.      | +1 315 268 1925 (9201 FAX)  | What is thee doing about it?
Potsdam, NY 13676 | Capitalists try to avoid a free market.  Why might that be?

2. X Windows with ET4000 PCI card ?

3. Kernel change summary 1.1.91 -> 1.1.92

4. RESULT: comp.security.firewalls passes 647:88

5. Problems with Kernel 1.1.93

6. 1.2.5 FPU won't compile

7. WD 1.08 Gig, Kernel 1.1.93

8. Want to do partition and then load Linux on the Win2K PC, how can I do that?

9. Problems with Kernel 1.1.93

10. Problems while compiling kernel 1.1.93 und 1.1.94

11. Kernel 1.1.89 & 1.1.92 ptrace vs 1.0.9 and ups debugger

12. FTAPE 2.02 with Kernel 1.1.92....


All times are UTC
Board index