- 05 10月, 2006 1 次提交
-
-
由 David Howells 提交于
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: NDavid Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
-
- 12 8月, 2006 1 次提交
-
-
由 Orjan Friberg 提交于
In my quest to try and figure out why test 14 (control write) doesn't work with my EZ-USB board, I noticed that sometimes testusb reported no error even though the kernel log complained "byte 0 is 0 not 2" etc. Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
- 01 7月, 2006 1 次提交
-
-
由 Jörn Engel 提交于
Signed-off-by: NJörn Engel <joern@wohnheim.fh-wedel.de> Signed-off-by: NAdrian Bunk <bunk@stusta.de>
-
- 22 6月, 2006 2 次提交
-
-
由 Alan Stern 提交于
This patch (as693b) makes the usbtest driver report errors in the isochronous bulk transfer tests instead of always returning 0. As an arbitrary cutoff, an error is returned if more than 10% of the packet transfers fail. It also stops a test immediately upon receiving an URB submission error. For a test harness, it's especially important to report when errors occur! Signed-off-by: NAlan Stern <stern@rowland.harvard.edu> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
由 Franck Bui-Huu 提交于
ctrl_complete functions acquires ctx->lock and tries to unlink all queued urbs in case of errors through usb_unlink_urb func. In its turn usb_unlink_urb calls, through the hcd driver, usb_hcd_giveback_urb which calls ctrl_complete again. At this time, ctx->lock is already taken by the same function. Signed-off-by: NFranck Bui-Huu <vagabon.xyz@gmail.com> Cc: David Brownell <david-b@pacbell.net> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
- 15 4月, 2006 1 次提交
-
-
由 David Brownell 提交于
Previously, scatterlist tests didn't write patterned data. Given how many corner cases are addresed by them, this was a significant gap in Linux-USB test coverage. Moreover, when peripherals checked for correct data patterns, false error reports would drown out the true ones. This adds the pattern on the way OUT from the host, so scatterlist tests can now be used to uncover bugs like host TX or peripheral RX paths failing for back-to-back short packets. It's easy enough to get an error there with at least one of the {DMA,PIO}{RX,TX} code paths, or run into hardware races that need to be defended against. Note this patch doesn't add checking for correct data patterns on the way IN from peripherals, just a FIXME for later. Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
- 21 3月, 2006 1 次提交
-
-
由 Eric Sesterhenn 提交于
Signed-off-by: NEric Sesterhenn <snakebyte@gmx.de> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
- 05 1月, 2006 1 次提交
-
-
由 Greg Kroah-Hartman 提交于
It is no longer needed, so let's remove it, saving a bit of memory. Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
- 18 11月, 2005 1 次提交
-
-
由 Greg Kroah-Hartman 提交于
This lets us remove a lot of code in the drivers that were all checking the same thing. It also found some bugs in a few of the drivers, which has been fixed up. Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de> Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
-
- 30 10月, 2005 1 次提交
-
-
由 David Hardeman 提交于
This patch uses sg_set_buf/sg_init_one in some places where it was duplicated. Signed-off-by: NDavid Hardeman <david@2gen.com> Cc: James Bottomley <James.Bottomley@steeleye.com> Cc: Greg KH <greg@kroah.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jeff Garzik <jgarzik@pobox.com> Signed-off-by: NAndrew Morton <akpm@osdl.org> Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
-
- 29 10月, 2005 2 次提交
-
-
由 Alan Stern 提交于
This patch (as582) adds a missing transfer_flags setting to the usbtest driver. Signed-off-by: NAlan Stern <stern@rowland.harvard.edu> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
由 David Brownell 提交于
This updates the handling of power state for USB interfaces. - Formalizes an existing invariant: interface "power state" is a boolean: ON when I/O is allowed, and FREEZE otherwise. It does so by defining some inlined helpers, then using them. - Adds a useful invariant: the only interfaces marked active are those bound to non-suspended drivers. Later patches build on this invariant. - Simplifies the interface driver API (and removes some error paths) by removing the requirement that they record power state changes during suspend and resume callbacks. Now usbcore does that. A few drivers were simplified to address that last change. Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de> drivers/usb/core/hub.c | 33 +++++++++------------ drivers/usb/core/message.c | 1 drivers/usb/core/usb.c | 65 +++++++++++++++++++++++++++++++++---------- drivers/usb/core/usb.h | 18 +++++++++++ drivers/usb/input/hid-core.c | 2 - drivers/usb/misc/usbtest.c | 10 ------ drivers/usb/net/pegasus.c | 2 - drivers/usb/net/usbnet.c | 2 - 8 files changed, 85 insertions(+), 48 deletions(-)
-
- 09 9月, 2005 1 次提交
-
-
由 Alan Stern 提交于
29 July 2005, Cambridge, MA: This afternoon Alan Stern submitted a patch to remove the URB_ASYNC_UNLINK flag from the Linux kernel. Mr. Stern explained, "This flag is a relic from an earlier, less-well-designed system. For over a year it hasn't been used for anything other than printing warning messages." An anonymous spokesman for the Linux kernel development community commented, "This is exactly the sort of thing we see happening all the time. As the kernel evolves, support for old techniques and old code can be jettisoned and replaced by newer, better approaches. Proprietary operating systems do not have the freedom or flexibility to change so quickly." Mr. Stern, a staff member at Harvard University's Rowland Institute who works on Linux only as a hobby, noted that the patch (labelled as548) did not update two files, keyspan.c and option.c, in the USB drivers' "serial" subdirectory. "Those files need more extensive changes," he remarked. "They examine the status field of several URBs at times when they're not supposed to. That will need to be fixed before the URB_ASYNC_UNLINK flag is removed." Greg Kroah-Hartman, the kernel maintainer responsible for overseeing all of Linux's USB drivers, did not respond to our inquiries or return our calls. His only comment was "Applied, thanks." Signed-off-by: NAlan Stern <stern@rowland.harvard.edu> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
- 05 9月, 2005 1 次提交
-
-
由 Pavel Machek 提交于
This adds type-checking to pm_message_t, so that people can't confuse it with int or u32. It also allows us to fix "disk yoyo" during suspend (disk spinning down/up/down). [We've tried that before; since that cpufreq problems were fixed and I've tried make allyes config and fixed resulting damage.] Signed-off-by: NPavel Machek <pavel@suse.cz> Signed-off-by: NAlexander Nyberg <alexn@telia.com> Signed-off-by: NAndrew Morton <akpm@osdl.org> Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
-
- 28 6月, 2005 1 次提交
-
-
由 David Brownell 提交于
Updates to "usbtest" driver: * Improve some diagnostics. One path that never generated diagnostics before should now generate two ... unless you hit a GCC bug that all my compilers seem to have, go figure. * Add suspend/resume support, so this behaves when the Linux host being used for testing suspends. * Don't test the "zero byte ep0 read" case unless real-world relevance for the testing is is irrelevant. Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net> Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
-
- 17 4月, 2005 1 次提交
-
-
由 Linus Torvalds 提交于
Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
-