From: Maxim Levitsky <maximlevitsky@gmail.com>
To: ipw3945-devel@lists.sourceforge.net
Cc: linux-wireless@vger.kernel.org,
"wally@theblackmoor.net" <wally@theblackmoor.net>
Subject: Re: Very big upload speed regression on iwl3945
Date: Sun, 07 Dec 2008 21:36:32 +0200 [thread overview]
Message-ID: <1228678592.20100.20.camel@maxim-laptop> (raw)
In-Reply-To: <1228678422.20100.17.camel@maxim-laptop>
Sorry, forgot to cc Wally Francis
On Sun, 2008-12-07 at 21:33 +0200, Maxim Levitsky wrote:
> Hi,
>
> I have spent whole day on this, but I am close.
> you, Walter Francis, were completely right that there is huge speed
> regression between 2.6.25 and 2.6.26, take look at the results:
>
> I am talking now about transfer speeds between router and iwl3945, so
> download speed is more or less the same on 2.6.26 and .25, it dropped
> from 2.4 Mb/s to 2.0 Mb/s, still a regression, but upload speed dropped
> from 3.0 (!) Mb/s to just 1.2 Mb/s maximum.
>
> Note, that I observed sharp drops in speed on both .26 and .25, after a
> while.
>
> on 2.6.25 speeds drops drasticly from 3.0 Mb/s /2.4 Mb/s to 1.0 Mb/s ,
> 150 Kb/s (not typo)
>
> on 2.6.26, upload speed changes a lot, and it is usually at 600~800
> Kb/s, but download speeds also drops to 150 kb/s.
> Those drops happen at same time (both upload and download).
>
>
> Bisect results:
>
>
> 17eed249539a7b756ca65a5cb0940abc48ef553b <- bad (1)
>
> 6002f266b3e7f0acc2d5158cddbed41730b02e82 <- doesn't load (2)
>
> 8318d78a44d49ac1edf2bdec7299de3617c4232e <- doesn't load (3)
> <cfg80211 APIfor channels/bitrates, mac80211 and driver conversion>
>
> 10b6b80145cc93887dd8aab99bfffa375e9add31 <- good (predicted) (4)
>
>
>
>
> (4) is the last good revision, then next revision (3) makes the driver
> show a backtrace, and fail.
>
> (2) still doesn't load
>
> (1) loads and have low transfer speeds.
>
> Currently I search for commit between (1) and (2) that fixed loading
> issue, then I will try to backport it, to find the exact commit.
>
>
>
> the backtrace:
>
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374773] [ext3:warn_on_slowpath+0x5f/0xc20] warn_on_slowpath+0x5f/0x90
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374781] [acpi_cpufreq:kmem_cache_alloc+0x6d/0x530] ? kmem_cache_alloc+0x6d/0xb0
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374785] [wiphy_register+0x14f/0x170] wiphy_register+0x14f/0x170
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374788] [iwl3945:ieee80211_register_hw+0x2d/0x330] ieee80211_register_hw+0x2d/0x2c0
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374799] [iwl3945:iwl3945_pci_probe+0xd0c/0x1690] :iwl3945:iwl3945_pci_probe+0xd0c/0x1690
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374803] [find_inode+0x53/0x90] ? find_inode+0x53/0x90
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374807] [sysfs_ilookup_test+0x0/0x20] ? sysfs_ilookup_test+0x0/0x20
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374810] [ext3:iput+0x4a/0x750] ? iput+0x4a/0x90
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374814] [sysfs_find_dirent+0x2d/0x40] ? sysfs_find_dirent+0x2d/0x40
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374819] [snd_pcm:_spin_lock+0x16/0x80] ? _spin_lock+0x16/0x40
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374822] [snd_pcm:_spin_unlock+0x10/0x100] ? _spin_unlock+0x10/0x40
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374826] [pci_device_probe+0x7c/0xa0] pci_device_probe+0x7c/0xa0
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374833] [driver_probe_device+0xa2/0x1e0] driver_probe_device+0xa2/0x1e0
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374836] [__driver_attach+0x0/0xb0] ? __driver_attach+0x0/0xb0
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374839] [__driver_attach+0xab/0xb0] __driver_attach+0xab/0xb0
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374843] [bus_for_each_dev+0x6b/0xa0] bus_for_each_dev+0x6b/0xa0
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374846] [acpi_cpufreq:kmem_cache_alloc+0x6d/0x530] ? kmem_cache_alloc+0x6d/0xb0
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374850] [usbcore:driver_attach+0x1c/0x3d0] driver_attach+0x1c/0x20
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374853] [bus_add_driver+0x1f8/0x270] bus_add_driver+0x1f8/0x270
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374857] [usbcore:driver_register+0x47/0x120] driver_register+0x47/0x120
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374860] [tg3:__pci_register_driver+0x5a/0x2b0] __pci_register_driver+0x5a/0xa0
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374867] [iwl3945:init_module+0x3f/0xb5000] :iwl3945:iwl3945_init+0x3f/0x93
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374871] [sys_init_module+0x154/0x1e90] sys_init_module+0x154/0x1e90
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374876] [snd_hda_intel:_spin_unlock_irqrestore+0x12/0xc0] ? _spin_unlock_irqrestore+0x12/0x40
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374881] [<ffffffff804be4b0>] ? ieee80211_rx_irqsafe+0x0/0x80
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374887] [system_call_after_swapgs+0x7b/0x80] system_call_after_swapgs+0x7b/0x80
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374890]
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374892] ---[ end trace 4e9e244a59472c97 ]---
> Dec 6 15:55:47 maxim-laptop kernel: [ 16.374894] iwl3945: Failed to register network device (error -22)
>
>
> Bes
>
prev parent reply other threads:[~2008-12-07 19:36 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-12-07 19:33 Very big upload speed regression on iwl3945 Maxim Levitsky
2008-12-07 19:36 ` Maxim Levitsky [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1228678592.20100.20.camel@maxim-laptop \
--to=maximlevitsky@gmail.com \
--cc=ipw3945-devel@lists.sourceforge.net \
--cc=linux-wireless@vger.kernel.org \
--cc=wally@theblackmoor.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox