Skip to content

K
Kernel

openeuler / Kernel
大约 1 年 前同步成功

通知 5
Star 0
Fork 0
K
Kernel

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 07 12月, 2006 1 次提交
  3. 05 12月, 2006 1 次提交
  5. 03 12月, 2006 1 次提交
  7. 22 11月, 2006 1 次提交
  9. 01 11月, 2006 1 次提交
  11. 05 10月, 2006 1 次提交
    • D
      IRQ: Maintain regs pointer globally rather than passing to IRQ handlers · 7d12e780
      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)
      7d12e780
  13. 23 9月, 2006 1 次提交
  15. 14 9月, 2006 1 次提交
  17. 06 9月, 2006 1 次提交
    • B
      [PATCH] myri10ge: improve firmware selection · ce7f9368
      由 Brice Goglin 提交于 
      Improve the firmware selection by adding 2 cases where we should use the
optimized firmware:
* when the actual PCIe link width is lower than 8x.
* when the board is plugged to one of the new Intel PCIe chipsets that
  are known to provide aligned PCIe completions.

The patch actually raises two concerns:
* We might want to add a generic PCI function to get the PCIe link width since
  some other drivers (at least ipath) do the same. But we probably do not want
  to add a new function for every PCIe capability. I will probably look at it
  and discuss it on linux-pci in the future.
* As requested during the submission, the PCI ids of chipsets that are known to
  provided aligned completion are defined in the myri10ge code. If we keep adding
  new ones, it might become better to move them to pciids.h.
  But, this sort of quirk to detect these chipsets are very specific to our NIC,
  I don't think it is worth moving it to the PCI core until somebody else really
  needs it.
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>
      ce7f9368
  19. 24 8月, 2006 2 次提交
  21. 20 8月, 2006 2 次提交
  23. 09 8月, 2006 1 次提交
  25. 04 8月, 2006 2 次提交
  27. 29 7月, 2006 1 次提交
  29. 13 7月, 2006 1 次提交
    • B
      [PATCH] myri10ge return value fix · 4c2248cc
      由 Brice Goglin 提交于 
      Andrew Morton wrote:
>   All these functions return error codes, and we're not checking them.  We
>   should.  So there's a patch which marks all these things as __must_check,
>   which causes around 1,500 new warnings.
>

The following patch fixes such a warning in myri10ge.

Check pci_enable_device() return value in myri10ge_resume().
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>
      4c2248cc
  31. 09 7月, 2006 1 次提交
  33. 06 7月, 2006 4 次提交
  35. 03 7月, 2006 1 次提交
  37. 23 6月, 2006 5 次提交
    • H
      [NET]: Merge TSO/UFO fields in sk_buff · 7967168c
      由 Herbert Xu 提交于 
      Having separate fields in sk_buff for TSO/UFO (tso_size/ufo_size) is not
going to scale if we add any more segmentation methods (e.g., DCCP).  So
let's merge them.

They were used to tell the protocol of a packet.  This function has been
subsumed by the new gso_type field.  This is essentially a set of netdev
feature bits (shifted by 16 bits) that are required to process a specific
skb.  As such it's easy to tell whether a given device can process a GSO
skb: you just have to and the gso_type field and the netdev's features
field.

I've made gso_type a conjunction.  The idea is that you have a base type
(e.g., SKB_GSO_TCPV4) that can be modified further to support new features.
For example, if we add a hardware TSO type that supports ECN, they would
declare NETIF_F_TSO | NETIF_F_TSO_ECN.  All TSO packets with CWR set would
have a gso_type of SKB_GSO_TCPV4 | SKB_GSO_TCPV4_ECN while all other TSO
packets would be SKB_GSO_TCPV4.  This means that only the CWR packets need
to be emulated in software.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      7967168c
    • H
      [NET]: Avoid allocating skb in skb_pad · 5b057c6b
      由 Herbert Xu 提交于 
      First of all it is unnecessary to allocate a new skb in skb_pad since
the existing one is not shared.  More importantly, our hard_start_xmit
interface does not allow a new skb to be allocated since that breaks
requeueing.

This patch uses pskb_expand_head to expand the existing skb and linearize
it if needed.  Actually, someone should sift through every instance of
skb_pad on a non-linear skb as they do not fit the reasons why this was
originally created.

Incidentally, this fixes a minor bug when the skb is cloned (tcpdump,
TCP, etc.).  As it is skb_pad will simply write over a cloned skb.  Because
of the position of the write it is unlikely to cause problems but still
it's best if we don't do it.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      5b057c6b
    • B
      [PATCH] myri10ge - drop workaround pci_save_state() disabling MSI · bfcbb008
      由 Brice Goglin 提交于 
      We don't need to restore the state right after saving it for later recovery
since commit 99dc804d (PCI: disable msi mode
in pci_disable_device) now prevents pci_save_state() from disabling MSI.
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>
      bfcbb008
    • B
      [PATCH] myri10ge - drop workaround for the missing AER ext cap on nVidia CK804 · 69eb887b
      由 Brice Goglin 提交于 
      We don't need to hardcode the AER capability of the nVidia CK804 chipset
anymore since commit cf34a8e0 (PCI: nVidia
quirk to make AER PCI-E extended capability visible) now makes sure that
this cap will be available to pci_find_ext_capability().
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NJeff Garzik <jeff@garzik.org>
      69eb887b
    • A
      [PATCH] myri10ge build fix · bec0e859
      由 Andrew Morton 提交于 
      Someone changed skb_linearize().

Cc: Brice Goglin <bgoglin@myri.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
      bec0e859
  39. 09 6月, 2006 1 次提交
    • B
      [PATCH] myri10ge update · b10c0668
      由 Brice Goglin 提交于 
      The following patch updates the myri10ge to 1.0.0, with the following changes:
* Switch to dma_alloc_coherent API.
* Avoid PCI burst when writing the firmware on chipset with unaligned completions.
* Use ethtool_op_set_tx_hw_csum instead of ethtool_op_set_tx_csum.
* Include linux/dma-mapping.h to bring DMA_32/64BIT_MASK on all architectures
  (was missing at least on alpha).
* Some typo and warning fixes.
Signed-off-by: NBrice Goglin <brice@myri.com>
Signed-off-by: NAndrew J. Gallatin <gallatin@myri.com>

 drivers/net/myri10ge/myri10ge.c |   57 +++++++++++++++++++-----------
 1 file changed, 37 insertions(+), 20 deletions(-)
Signed-off-by: NJeff Garzik <jeff@garzik.org>
      b10c0668
  41. 24 5月, 2006 1 次提交