Most recent kernel where this bug did *NOT* occur: 2.6.21-rc7 Distribution: none Hardware Environment: i386, Pentium M, Sony laptop Software Environment: n/a Problem Description: n/a Steps to reproduce: n/a Patches against v2.6.21-rc7 will follow.
Created attachment 11238 [details] [PATCH 1/2] [mmc] alternative TI FM MMC/SD driver for 2.6.21-rc7 The device is present in many notebooks. Notebooks depend heavily on suspend/resume functionality. tifm_core/7xx1/sd family is an ambitous, but uncompleted project. It used to crash on resuming, or hang up on suspending. A less common failure used to be trigerred by a fast card insert/removal sequence. Finally, tifm_sd module needs to be manually inserted. This driver is kind of mutant. The bones are taken from sdhci and omap, the meat - from tifm_*. It contains all features (and bugs except named above) of tifm_* as it was in kernel 2.6.21-rc7. I have been testing this version since linux-2.6.18 (daily reading photos from cards, daily suspending/resuming) without a single glitch. This patch only provides sources. Kernel configuration follows in next message.
Created attachment 11239 [details] [PATCH 2/2] [mmc] alternative TIFM driver config for 2.6.21-rc7 * [TIFM_CORE] Updated prompt "...interface" -> "..virtual bus", updated help message to reflect implementation details. * [TIFM_7XX1] Updated help message; * [MMC_TIFM] New symbol to allow alternative. * [MMC_TIFM_7XX1] New symbol for the driver. This patch will preserve any existing kernel config concerning tifm modules. The agregate symbol [MMC_TIFM] describes [MMC_TIFM_7XX1], because the latter can be hidden if [TIFM_7XX1] is selected. [TIFM_7XX1] and [MMC_TIFM_7XX1] are mutually exclusive, but belong to different dirs. For that reason, we cannot use obvious 'choice' kernel config keyword. [MMC_TIFM_SD] and [MMC_TIFM_7XX1] are not mutually exclusive. If there is a virtual bus controller different that [TIFM_7XX1], those two modules can serve different cards at the same time.
Uploaded version compiles cleanly and runs smoothly on my sony laptop. I routinely keep the computer running for several weeks without reboots, but with daily suspend/resume. My photo uses mmc/sd cards, so it is stupid to use external card reader when you have a built-in one.
Created attachment 11240 [details] Unit test script [USES SUDO!] This is a "test suite" for those who like XP programming style, as I do.
Created attachment 11272 [details] patch removes custom debug macro
Created attachment 11273 [details] patch reduces delay on insert/removal
Created attachment 11308 [details] updated Unit test script [USES SUDO!] Reconfigured for [tifm_*]. All magic values replaced with variables.
Today's tar on http://download.berlios.de/tifmxx/tifm-0.8e.tar.bz2 seems to be stable enough. *** This bug has been marked as a duplicate of 8052 ***