The "iic" instance does not depend on i2c_bitbang, so it should not
express it as a dependency. That's the responsibility of the back-ends
that do require it.
Don't bother with the "no disk label" message. It's not particularly
useful, and is downright confusing if the disk has some other sort of
partition scheme (such at GPT).
t_dlclose_thread: Make timeouts manifest more predictably.
And use SIGABRT to get stack traces.
PR lib/59751: dlclose is not MT-safe depending on the libraries
unloaded
SCSI-2 and later define READ(6) as optional and READ(10) as mandatory
and some devices (e.g. the Linux iSCSI tgtd) reject the small commands.
MODE_SENSE(6) and MODE_SENSE(10) can both be optional, so it's possible
that a SCSI-2 device only implements READ(10) and MODE_SENSE(6), which
then needs further refinement.
t_dlclose_thread: Test recursive dlopen/dlclose too.
Simple test first:
- h_helper_recurdso dlopens h_helper_dso2
- h_helper_dso2 needs h_helper_dso1
When dlopening h_helper_recurdso, the initialization order must be:
ENTER h_helper_recurdso
-> h_helper_dso1
-> h_helper_dso2
LEAVE h_helper_recurdso
For the more complex test, we have the relations:
- h_helper_recurdso2 needs h_helper_recurdso
- h_helper_recurdso dlopens h_helper_dso2
- h_helper_dso2 needs h_helper_dso1
[17 lines not shown]
Add support for experimental Rendition Verite 3D acceleration interface.
This allow userland process to load 3D firmware (like v2000gl.uc) and
execute it on the chip's RISC CPU. It also provides the necessary support
infrastructure like basic VRAM accounting and ability to feed commands
and data through DMA.
In conjuction with customized userland Mesa, this allows OpenGL 1.x
workloads to run accelerated on the Verite.
While here, add related debugging facilities.