Firmware 4.33.19608

[NiLuJe]

Okay, on the Sage, it's much easier to reproduce with the NickelMenu Home menu, especially *after* closing the Battery popup.

[pid  1295] 15:42:13.363289 [b1bfd4ec] ioctl(6</dev/disp>, DISP_EINK_UPDATE2, {area={x_top=676, y_top=144, x_bottom=1396, y_bottom=558}, layer_num=1, update_mode=EINK_PARTIAL_MODE|EINK_GL16_MODE, lyr_cfg2={info={mode=LAYER_MODE_BUFFER, zorder=0, alpha_mode=1, alpha_value=255, screen_win={x=0, y=0, width=1440, height=1920}, b_trd_out=0, out_trd_mode=DISP_3D_OUT_MODE_TB, fb={fd=5, y8_fd=0, size=[{width=1440, height=1920}, {width=1440, height=1920}, {width=1440, height=1920}], align=[2, 2, 2], format=DISP_FORMAT_ARGB_8888, color_space=DISP_UNDEF, trd_right_fd=0, pre_multiply=1, crop={x=0, y=0, width=1440 << 32, height=1920 << 32}, flags=DISP_BF_NORMAL, scan=DISP_SCAN_PROGRESSIVE, eotf=DISP_EOTF_RESERVED, depth=0, fbd_en=0, metadata_fd=0, metadata_size=0, metadata_flag=0}, id=0, atw={used=0, mode=NORMAL_MODE, b_row=0, b_col=0, cof_fd=0}}, enable=1, channel=0, layer_id=0}, frame_id=[343], rotate=90, cfa_use=*0}) = 343 <0.025538>
[pid  1295] 15:42:13.408816 [b1bfd4ec] ioctl(6</dev/disp>, DISP_EINK_UPDATE2, {area={x_top=1056, y_top=0, x_bottom=1145, y_bottom=138}, layer_num=1, update_mode=EINK_PARTIAL_MODE|EINK_GL16_MODE, lyr_cfg2={info={mode=LAYER_MODE_BUFFER, zorder=0, alpha_mode=1, alpha_value=255, screen_win={x=0, y=0, width=1440, height=1920}, b_trd_out=0, out_trd_mode=DISP_3D_OUT_MODE_TB, fb={fd=5, y8_fd=0, size=[{width=1440, height=1920}, {width=1440, height=1920}, {width=1440, height=1920}], align=[2, 2, 2], format=DISP_FORMAT_ARGB_8888, color_space=DISP_UNDEF, trd_right_fd=0, pre_multiply=1, crop={x=0, y=0, width=1440 << 32, height=1920 << 32}, flags=DISP_BF_NORMAL, scan=DISP_SCAN_PROGRESSIVE, eotf=DISP_EOTF_RESERVED, depth=0, fbd_en=0, metadata_fd=0, metadata_size=0, metadata_flag=0}, id=0, atw={used=0, mode=NORMAL_MODE, b_row=0, b_col=0, cof_fd=0}}, enable=1, channel=0, layer_id=0}, frame_id=[344], rotate=90, cfa_use=*0}) = 344 <0.015941>
[pid  1295] 15:42:13.702241 [b1bfd4ec] ioctl(6</dev/disp>, DISP_EINK_UPDATE2, {area={x_top=1152, y_top=1781, x_bottom=1439, y_bottom=1919}, layer_num=1, update_mode=EINK_PARTIAL_MODE|EINK_GL16_MODE, lyr_cfg2={info={mode=LAYER_MODE_BUFFER, zorder=0, alpha_mode=1, alpha_value=255, screen_win={x=0, y=0, width=1440, height=1920}, b_trd_out=0, out_trd_mode=DISP_3D_OUT_MODE_TB, fb={fd=5, y8_fd=0, size=[{width=1440, height=1920}, {width=1440, height=1920}, {width=1440, height=1920}], align=[2, 2, 2], format=DISP_FORMAT_ARGB_8888, color_space=DISP_UNDEF, trd_right_fd=0, pre_multiply=1, crop={x=0, y=0, width=1440 << 32, height=1920 << 32}, flags=DISP_BF_NORMAL, scan=DISP_SCAN_PROGRESSIVE, eotf=DISP_EOTF_RESERVED, depth=0, fbd_en=0, metadata_fd=0, metadata_size=0, metadata_flag=0}, id=0, atw={used=0, mode=NORMAL_MODE, b_row=0, b_col=0, cof_fd=0}}, enable=1, channel=0, layer_id=0}, frame_id=[345], rotate=90, cfa_use=*0}) = 345 <0.030606>
[pid  1295] 15:42:13.953617 [b1bfd4ec] ioctl(6</dev/disp>, DISP_EINK_UPDATE2, {area={x_top=992, y_top=1139, x_bottom=1439, y_bottom=1781}, layer_num=1, update_mode=EINK_PARTIAL_MODE|EINK_GL16_MODE, lyr_cfg2={info={mode=LAYER_MODE_BUFFER, zorder=0, alpha_mode=1, alpha_value=255, screen_win={x=0, y=0, width=1440, height=1920}, b_trd_out=0, out_trd_mode=DISP_3D_OUT_MODE_TB, fb={fd=5, y8_fd=0, size=[{width=1440, height=1920}, {width=1440, height=1920}, {width=1440, height=1920}], align=[2, 2, 2], format=DISP_FORMAT_ARGB_8888, color_space=DISP_UNDEF, trd_right_fd=0, pre_multiply=1, crop={x=0, y=0, width=1440 << 32, height=1920 << 32}, flags=DISP_BF_NORMAL, scan=DISP_SCAN_PROGRESSIVE, eotf=DISP_EOTF_RESERVED, depth=0, fbd_en=0, metadata_fd=0, metadata_size=0, metadata_flag=0}, id=0, atw={used=0, mode=NORMAL_MODE, b_row=0, b_col=0, cof_fd=0}}, enable=1, channel=0, layer_id=0}, frame_id=[346], rotate=90, cfa_use=*0}) = 346 <0.026500>
[pid  1295] 15:42:13.982064 [b1bfd4ec] ioctl(6</dev/disp>, DISP_EINK_UPDATE2, {area={x_top=1152, y_top=1782, x_bottom=1439, y_bottom=1919}, layer_num=1, update_mode=EINK_PARTIAL_MODE|EINK_GL16_MODE, lyr_cfg2={info={mode=LAYER_MODE_BUFFER, zorder=0, alpha_mode=1, alpha_value=255, screen_win={x=0, y=0, width=1440, height=1920}, b_trd_out=0, out_trd_mode=DISP_3D_OUT_MODE_TB, fb={fd=5, y8_fd=0, size=[{width=1440, height=1920}, {width=1440, height=1920}, {width=1440, height=1920}], align=[2, 2, 2], format=DISP_FORMAT_ARGB_8888, color_space=DISP_UNDEF, trd_right_fd=0, pre_multiply=1, crop={x=0, y=0, width=1440 << 32, height=1920 << 32}, flags=DISP_BF_NORMAL, scan=DISP_SCAN_PROGRESSIVE, eotf=DISP_EOTF_RESERVED, depth=0, fbd_en=0, metadata_fd=0, metadata_size=0, metadata_flag=0}, id=0, atw={used=0, mode=NORMAL_MODE, b_row=0, b_col=0, cof_fd=0}}, enable=1, channel=0, layer_id=0}, frame_id=[347], rotate=90, cfa_use=*0}) = 347 <0.034047>
[pid  1295] 15:42:14.018351 [b1bfd4ec] ioctl(6</dev/disp>, DISP_EINK_WAIT_FRAME_SYNC_COMPLETE, 347) = 0 <0.376256>

It also happens with long presses on cover thumbnails in the Home screen's top row (e.g., closing the battery popup, and long-press on a thumbnail that was below the popup), so I'm guessing it has to do with races/collision handling, possibly because mixed content is slower to process...

Shit happens during the DISP_EINK_WAIT_FRAME_SYNC_COMPLETE in every case (and that appears to be called by the floating menu code, hence the NM menu & long-press menu triggers).

EDIT: Strangely enough, on the Forma, while the cover thumbnail menu does bracket everything in a couple of MXCFB_WAIT_FOR_UPDATE_COMPLETE_V3 ioctls, the NM menu does NOT.

[Deobulakenyo]

On Libra 2, i tested the 24-line spacing patch and within the hour, freezing/hanging while reading a kepub.

Flashed a bunch of patches with removed line sacing patch and no freezing.

[Lykke]

Originally Posted by Deobulakenyo

On Libra 2, i tested the 24-line spacing patch and within the hour, freezing/hanging while reading a kepub.

Flashed a bunch of patches with removed line sacing patch and no freezing.

I have applied the same patch. If I have another freeze I will try removing it.

[Sirtel]

Originally Posted by Deobulakenyo

On Libra 2, i tested the 24-line spacing patch and within the hour, freezing/hanging while reading a kepub.

Flashed a bunch of patches with removed line sacing patch and no freezing.

I haven't had a freeze either on my Sage or my Libra 2 with the same patch (4.32). And I have read some large books on both of them.

[geek1011]

Originally Posted by Deobulakenyo

On Libra 2, i tested the 24-line spacing patch and within the hour, freezing/hanging while reading a kepub.

Flashed a bunch of patches with removed line sacing patch and no freezing.

That doesn't make sense, since the only thing that patch effectively changes which could have any effect is the number of times QList<double>::append is called.

[Deobulakenyo]

Originally Posted by geek1011

That doesn't make sense, since the only thing that patch effectively changes which could have any effect is the number of times QList<double>::append is called.

A lot of things do not make sense with Libra 2.

[windirt]

I installed nanoclock 0.11.4 and set the following two configurations as per @Deobulakenyo's ini configuration

backgroundless=false
overlay=false

In the past night of use, I still had two crashes.
Libra 2, 4.33, nanoclock 0.11.4

enabled patches:

Un-Force user text-align in div,p tags in KePubs: yes
Un-Force user font-family in KePubs: yes
Un-force link decoration in KePubs: yes
Reduce top/bottom page spacer: yes
Custom synopsis details line spacing: yes
Custom synopsis font size: yes
Increase home screen cover size: yes
Dictionary pop-up - increase available text area: yes
Increase library cover size: yes
Reduce new header/footer height: yes
Show all games: yes

After uninstalling nanoclock, no crashes have occurred so far.

[geek1011]

Originally Posted by windirt

I installed nanoclock 0.11.4 and set the following two configurations as per @Deobulakenyo's ini configuration


In the past night of use, I still had two crashes.
Libra 2, 4.33, nanoclock 0.11.4

enabled patches:


After uninstalling nanoclock, no crashes have occurred so far.

It's probably nanoclock then, since those patches have next to no chance of causing issues (most are CSS patches, and all are trivial).

[Deobulakenyo]

Originally Posted by windirt

I installed nanoclock 0.11.4 and set the following two configurations as per @Deobulakenyo's ini configuration


In the past night of use, I still had two crashes.
Libra 2, 4.33, nanoclock 0.11.4

enabled patches:


After uninstalling nanoclock, no crashes have occurred so far.

On my end i set to true both the overlay and background using nanoclock 0.11.4 and no freezing whatsoever on my Libra 2 unless i use the 24-lines spacing patch.

Weir, weird, weird

[JSWolf]

Originally Posted by Deobulakenyo

On my end i set to true both the overlay and background using nanoclock 0.11.4 and no freezing whatsoever on my Libra 2 unless i use the 24-lines spacing patch.

Weir, weird, weird

Remove nanoclock and enable the 24-line spacing patch and see how you get on.

[Deobulakenyo]

Originally Posted by JSWolf

Remove nanoclock and enable the 24-line spacing patch and see how you get on.

I need nanoclock. With nanoclock and without the line spacing patches, everything is fine.

[JSWolf]

Originally Posted by Deobulakenyo

I need nanoclock. With nanoclock and without the line spacing patches, everything is fine.

Nobody needs nanoclock.

You can reinstall after doing this test. You are the only one posting of this issue so you are the best one to do the test.

[Deobulakenyo]

Originally Posted by JSWolf

Nobody needs nanoclock.

You can reinstall after doing this test. You are the only one posting of this issue so you are the best one to do the test.

Will do it when i get to it. Probably during lunch break tomorrow

[rashkae]

If the speculation is correct that there is some instability related to processor weirdness when clock speed is too low, there can easily be a great variety of individually affected units and trigger conditions. Someone will probably have to create some easy to apply patches for people to test that theory.

[JSWolf]

Originally Posted by Deobulakenyo

Will do it when i get to it. Probably during lunch break tomorrow

Thank you.