Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit) |
||
---|---|---|
.. | ||
compressor | ||
lowlevel | ||
zftape | ||
Kconfig | ||
Makefile | ||
README.PCI | ||
RELEASE-NOTES |
README.PCI
Some notes for ftape users with PCI motherboards: ================================================= The problem: ------------ There have been some problem reports from people using PCI-bus based systems getting overrun errors. I wasn't able to reproduce these until I ran ftape on a Intel Plato (Premiere PCI II) motherboard with bios version 1.00.08AX1. It turned out that if GAT (Guaranteed Access Timing) is enabled (?) ftape gets a lot of overrun errors. The problem disappears when disabling GAT in the bios. Note that Intel removed this setting (permanently disabled) from the 1.00.10AX1 bios ! It looks like that if GAT is enabled there are often large periods (greater than 120 us !??) on the ISA bus that the DMA controller cannot service the floppy disk controller. I cannot imagine this being acceptable in a decent PCI implementation. Maybe this is a `feature' of the chipset. I can only speculate why Intel choose to remove the option from the latest Bios... The lesson of this all is that there may be other motherboard implementations having the same of similar problems. If you experience a lot of overrun errors during a backup to tape, see if there is some setting in the Bios that may influence the bus timing. I judge this a hardware problem and not a limitation of ftape ;-) My DOS backup software seems to be suffering from the same problems and even refuses to run at 1 Mbps ! Ftape will reduce the data-rate from 1 Mbps to 500 Kbps if the number of overrun errors on a track exceeds a threshold. Possible solutions: ------------------- Some of the problems were solved by upgrading the (flash) bios. Other suggest that it has to do with the FDC being on the PCI bus, but that is not the case with the Intel Premiere II boards. [If upgrading the bios doesn't solve the problem you could try a floppy disk controller on the isa-bus]. Here is a list of systems and recommended BIOS settings: Intel Premiere PCI (Revenge): Bios version 1.00.09.AF2 is reported to work. Intel Premiere PCI II (Plato): Bios version 1.00.10.AX1 and version 11 beta are ok. If using version 1.00.08.AX1, GAT must be disabled ! ASUS PCI/I-SP3G: Preferred settings: ISA-GAT-mode : disabled DMA-linebuffer-mode : standard ISA-masterbuffer-mode : standard DELL Dimension XPS P90 Bios version A2 is reported to be broken, while bios version A5 works. You can get a flash bios upgrade from http://www.dell.com To see if you're having the GAT problem, try making a backup under DOS. If it's very slow and often repositions you're probably having this problem. --//-- LocalWords: ftape PCI bios GAT ISA DMA chipset Mbps Kbps FDC isa AF ok ASUS LocalWords: SP linebuffer masterbuffer XPS http www com