Recent Posts
91
Repairing and Upgrading Rockbox Capable Players / Re: SSD sold by iPod Parts UK
« Last post by Frankenpod on April 18, 2025, 11:40:23 AM »Confusingly, with 1TB, 400GB, 400Gb, 400GB cards, it worked one time I did it and came out as 2032GiB, yet the second time I tried the same combo of cards it wouldn't work. 2200GB seems to translate to 2048.9GiB, which would surely be almost exactly 2TiB, so presumably would be right on the margin for exceeding the limit? Though then don't know why the one that worked came out that far below 2048GiB.
Maybe some cards of the same nominal size are not actually the full stated capacity? I dunno, I don't understand much as to how this all works.
Maybe some cards of the same nominal size are not actually the full stated capacity? I dunno, I don't understand much as to how this all works.
92
Repairing and Upgrading Rockbox Capable Players / Re: SSD sold by iPod Parts UK
« Last post by Frankenpod on April 18, 2025, 11:29:48 AM »The last one I tried >2TiB was an iFlash quad with 1500, 512, 128, 64 cards. Had to replace the 64 with a 32 for it to work. With the 64 in there it would restore (with iTunes under windows) as 27GiB (!). Which is pretty consistent with what happened with all my previous experiments with exceeding 2TiB.
With the 32GB card instead it restored OK and rockbox reports it as 2045GiB (which I guess is a smidgeon below the 2TiB, or 2048GiB, limit, so probably the best one can hope for).
Didn't try the first one with Linux, Windows definitely could only see 27GiB, no matter what way I tried to look at the drive, nor did I try anything clever with putting the iFlash in some other kind of adaptor, only in the iPod (not at all sure what the computer would see if one did that, given that the ipod, to the limited degree I understand it, does some weirdness that obfuscates what the drive actually is and presents something quasi-fictional to the computer?).
Maybe will give that a go the next one I build.
With the 32GB card instead it restored OK and rockbox reports it as 2045GiB (which I guess is a smidgeon below the 2TiB, or 2048GiB, limit, so probably the best one can hope for).
Didn't try the first one with Linux, Windows definitely could only see 27GiB, no matter what way I tried to look at the drive, nor did I try anything clever with putting the iFlash in some other kind of adaptor, only in the iPod (not at all sure what the computer would see if one did that, given that the ipod, to the limited degree I understand it, does some weirdness that obfuscates what the drive actually is and presents something quasi-fictional to the computer?).
Maybe will give that a go the next one I build.
93
Repairing and Upgrading Rockbox Capable Players / Re: SSD sold by iPod Parts UK
« Last post by speachy on April 18, 2025, 10:51:14 AM »Have been experimenting with getting the closest I can to the 2TiB limit. 1TB+3x400GB or 1500+512+128+32 seem to be the closest one can get without going over that limit (if you exceed it by even a few GB it won't restore properly). Significantly 2TiB is not the same as 2TB - that confused me for a while.
I'd be curious to see what >2TiB on one of these adapters would result in when looked at on a PC running Linux. (there are ZIF-to-ATA adapters out there)
(Just want to make sure that if it doesn't work, it's due to the SD adapter being braindead, rather than a problem with Rockbox itself)
94
Repairing and Upgrading Rockbox Capable Players / Re: SSD sold by iPod Parts UK
« Last post by Frankenpod on April 18, 2025, 09:42:48 AM »You aren't wrong that iFlash adaptors can be troublesome.
I have been using two iPods every day for several months- Classic 7 and Video 5.5, they both work with flash-Quad, one with 256Gb Samsung and the other with Kingston 256Gb. I have never noticed any problems or errors related to the adapter. Everything works fine, used with the latest version of RockBox
Been faffing around with them for over a decade now. Don't want to mention how many I currently have, it's an embarrassingly excessive number (the majority of them 7.5gens). They do work eventually, but had so many weird issues with them over the years. In particular certain combos of cards (different brands and capacities) will sometimes refuse to work together in the same ipod, but work fine if you rearrange them in different permutations in different ipods. The sheer variety of different weird failure modes you can encounter is amazing to me.
Have been experimenting with getting the closest I can to the 2TiB limit. 1TB+3x400GB or 1500+512+128+32 seem to be the closest one can get without going over that limit (if you exceed it by even a few GB it won't restore properly). Significantly 2TiB is not the same as 2TB - that confused me for a while.
95
Repairing and Upgrading Rockbox Capable Players / Re: SSD sold by iPod Parts UK
« Last post by EHOT on April 18, 2025, 07:41:24 AM »You aren't wrong that iFlash adaptors can be troublesome.
I have been using two iPods every day for several months- Classic 7 and Video 5.5, they both work with flash-Quad, one with 256Gb Samsung and the other with Kingston 256Gb. I have never noticed any problems or errors related to the adapter. Everything works fine, used with the latest version of RockBox
96
Starting Development and Compiling / Re: Build Failing...
« Last post by bahus on April 18, 2025, 06:39:50 AM »Also I used 247 and not 249 when selecting the build target.
249) Eros Q / K native (hw4 bl only)
My guess "hw4 bl only" means Bootloader only build for hw4 variant. So, no wonder it fails for Normal build.
97
Starting Development and Compiling / Re: Build Failing...
« Last post by MinorKey on April 18, 2025, 06:06:39 AM »I finally got it to work. Not sure what was wrong before, but I cleared the TMP directory and ran rockboxdev.sh again and made sure that that finished properly.
Also I used 247 and not 249 when selecting the build target. In hindsight that could have been the whole problem. Either way, cool that it's finally working.
Also I used 247 and not 249 when selecting the build target. In hindsight that could have been the whole problem. Either way, cool that it's finally working.
98
Starting Development and Compiling / Re: Build Failing...
« Last post by MinorKey on April 18, 2025, 03:52:36 AM »Thank you for your reply.
I think I posted the wrong output. Well not entirely wrong, but that's the output generated after running make -j for the second time, when there's already something in the build directory. This is actually what's generated generated after running make -j when that dir is blank:
CC lib/rbcodec/metadata/oma.c
CC lib/rbcodec/metadata/rm.c
CC lib/rbcodec/metadata/sgc.c
CC lib/rbcodec/metadata/sid.c
CC lib/rbcodec/metadata/smaf.c
CC lib/rbcodec/metadata/spc.c
CC lib/rbcodec/metadata/tta.c
CC lib/rbcodec/metadata/vgm.c
CC lib/rbcodec/metadata/vorbis.c
CC lib/rbcodec/metadata/vox.c
CC lib/rbcodec/metadata/wave.c
CC lib/rbcodec/metadata/wavpack.c
CC lib/rbcodec/metadata/aac.c
CC lib/mipsunwinder/backtrace-mips32.c
CC lib/mipsunwinder/init_context_32.S
CC lib/skin_parser/skin_buffer.c
CC lib/skin_parser/skin_parser.c
CC lib/skin_parser/skin_scan.c
CC lib/skin_parser/tag_table.c
CC lib/fixedpoint/fixedpoint.c
make: *** No rule to make target '/home/matej/git/rockbox/firmware/export/config/erosqnative_v4.h', needed by '/home/matej/git/rockbox/build/ram.link'. Stop.
make: *** Waiting for unfinished jobs....
matej@Linux-PC:~/git/rockbox/build$
It's as if the make process is hanging or something, but of course, I'm no expert when it comes to this.
Also this is what the configure step tells me after setting my target, but probably this is all OK:
229) NWZ-S750 series 233) DX90 240) Rocker
==FiiO== ==xDuoo== ==AIGO==
244) M3K Linux 241) X3 245) Eros Q / K
246) M3K baremetal 242) X3II 247) Eros Q / K native
243) X20 (hw1/hw2 bl, all hw rb)
==Shanling== 248) Eros Q / K native
260) Q1 (hw3 bl only)
249) Eros Q / K native
(hw4 bl only)
249
Platform set to erosqnative_v4
Build (N)ormal, (A)dvanced, (S)imulator, (B)ootloader, (C)heckWPS, (D)atabase tool, (W)arble codec tool: (Defaults to N)
n
Normal build selected
Using source code root directory: /home/matej/git/rockbox
Using mipsel-elf-gcc 4.9.4 (409)
Using mipsel-elf-ld
Detected make GNU Make 4.3
Found and uses ccache (/usr/bin/ccache)
Automatically selected arch: mips (ver 32)
Created Makefile
matej@Linux-PC:~/git/rockbox/build$
Edit: More/Fixed info
I think I posted the wrong output. Well not entirely wrong, but that's the output generated after running make -j for the second time, when there's already something in the build directory. This is actually what's generated generated after running make -j when that dir is blank:
CC lib/rbcodec/metadata/oma.c
CC lib/rbcodec/metadata/rm.c
CC lib/rbcodec/metadata/sgc.c
CC lib/rbcodec/metadata/sid.c
CC lib/rbcodec/metadata/smaf.c
CC lib/rbcodec/metadata/spc.c
CC lib/rbcodec/metadata/tta.c
CC lib/rbcodec/metadata/vgm.c
CC lib/rbcodec/metadata/vorbis.c
CC lib/rbcodec/metadata/vox.c
CC lib/rbcodec/metadata/wave.c
CC lib/rbcodec/metadata/wavpack.c
CC lib/rbcodec/metadata/aac.c
CC lib/mipsunwinder/backtrace-mips32.c
CC lib/mipsunwinder/init_context_32.S
CC lib/skin_parser/skin_buffer.c
CC lib/skin_parser/skin_parser.c
CC lib/skin_parser/skin_scan.c
CC lib/skin_parser/tag_table.c
CC lib/fixedpoint/fixedpoint.c
make: *** No rule to make target '/home/matej/git/rockbox/firmware/export/config/erosqnative_v4.h', needed by '/home/matej/git/rockbox/build/ram.link'. Stop.
make: *** Waiting for unfinished jobs....
matej@Linux-PC:~/git/rockbox/build$
It's as if the make process is hanging or something, but of course, I'm no expert when it comes to this.
Also this is what the configure step tells me after setting my target, but probably this is all OK:
229) NWZ-S750 series 233) DX90 240) Rocker
==FiiO== ==xDuoo== ==AIGO==
244) M3K Linux 241) X3 245) Eros Q / K
246) M3K baremetal 242) X3II 247) Eros Q / K native
243) X20 (hw1/hw2 bl, all hw rb)
==Shanling== 248) Eros Q / K native
260) Q1 (hw3 bl only)
249) Eros Q / K native
(hw4 bl only)
249
Platform set to erosqnative_v4
Build (N)ormal, (A)dvanced, (S)imulator, (B)ootloader, (C)heckWPS, (D)atabase tool, (W)arble codec tool: (Defaults to N)
n
Normal build selected
Using source code root directory: /home/matej/git/rockbox
Using mipsel-elf-gcc 4.9.4 (409)
Using mipsel-elf-ld
Detected make GNU Make 4.3
Found and uses ccache (/usr/bin/ccache)
Automatically selected arch: mips (ver 32)
Created Makefile
matej@Linux-PC:~/git/rockbox/build$
Edit: More/Fixed info
99
Starting Development and Compiling / Re: Build Failing...
« Last post by speachy on April 17, 2025, 09:25:07 PM »I honestly have no idea what's wrong, other than perhaps something failing earlier in the build leading to this, most likely due to something being different from one environment/system to the other.
After every commit we rebuild _everything_ -- all binary packages, bootloaders, checkwpses, simulators, and manuals (nearly 400 separate builds in total!) and the board is currently completely warning-free for all of 'em.
(see https://build.rockbox.org/dev.cgi )
After every commit we rebuild _everything_ -- all binary packages, bootloaders, checkwpses, simulators, and manuals (nearly 400 separate builds in total!) and the board is currently completely warning-free for all of 'em.
(see https://build.rockbox.org/dev.cgi )
100
Starting Development and Compiling / Build Failing...
« Last post by MinorKey on April 17, 2025, 11:32:58 AM »Hello,
I'm trying to set up my dev environment on a PC running Linux, but I'm getting an ugly set of errors and I have absolutely no clue what's going on. I had absolutely no problems getting it to work in the same distro on a virtual machine, so it's very odd that it refuses to do the same running on a real one for some reason.
I've downloaded the rockbox repo to a subfolder inside my home directory, but the result is the same even if I download it directly to my home folder.
This is what the terminal tells me after running make -j:
};
^
/home/matej/git/rockbox/apps/settings_list.c:114:40: warning: missing initializer for field ‘({anonymous})’ of ‘const struct settings_list’ [-Wmissing-field-initializers]
{.bool_setting=(struct bool_setting[]){{cb,yes_id,no_id,cfgvals}}} }
^
/home/matej/git/rockbox/apps/settings_list.c:118:13: note: in expansion of macro ‘BOOL_SETTING’
BOOL_SETTING(flags,var,lang_id,default,name,off_on, \
^
/home/matej/git/rockbox/apps/settings_list.c:2317:5: note: in expansion of macro ‘OFFON_SETTING’
OFFON_SETTING(0, playback_log, LANG_LOGGING, false, "play log", NULL),
^
In file included from /home/matej/git/rockbox/apps/settings_list.c:36:0:
/home/matej/git/rockbox/apps/settings_list.h:187:5: note: ‘({anonymous})’ declared here
};
^
make: *** [/home/matej/git/rockbox/tools/root.make:484: /home/matej/git/rockbox/build/apps/settings_list.o] Error 1
matej@Linux-PC:~/git/rockbox/build$
Here I was trying to build RB for theEros Q Native, but what I choose to build doesn't seem to make a difference.
I can still use my VM should it be necessary to test/update translations, but getting it to work on a real machine instead and getting rid of the VM would be far better.
So hope there's a solution for this, but be as it may, thanks for reading and have a nice day. 🙂
I'm trying to set up my dev environment on a PC running Linux, but I'm getting an ugly set of errors and I have absolutely no clue what's going on. I had absolutely no problems getting it to work in the same distro on a virtual machine, so it's very odd that it refuses to do the same running on a real one for some reason.
I've downloaded the rockbox repo to a subfolder inside my home directory, but the result is the same even if I download it directly to my home folder.
This is what the terminal tells me after running make -j:
};
^
/home/matej/git/rockbox/apps/settings_list.c:114:40: warning: missing initializer for field ‘({anonymous})’ of ‘const struct settings_list’ [-Wmissing-field-initializers]
{.bool_setting=(struct bool_setting[]){{cb,yes_id,no_id,cfgvals}}} }
^
/home/matej/git/rockbox/apps/settings_list.c:118:13: note: in expansion of macro ‘BOOL_SETTING’
BOOL_SETTING(flags,var,lang_id,default,name,off_on, \
^
/home/matej/git/rockbox/apps/settings_list.c:2317:5: note: in expansion of macro ‘OFFON_SETTING’
OFFON_SETTING(0, playback_log, LANG_LOGGING, false, "play log", NULL),
^
In file included from /home/matej/git/rockbox/apps/settings_list.c:36:0:
/home/matej/git/rockbox/apps/settings_list.h:187:5: note: ‘({anonymous})’ declared here
};
^
make: *** [/home/matej/git/rockbox/tools/root.make:484: /home/matej/git/rockbox/build/apps/settings_list.o] Error 1
matej@Linux-PC:~/git/rockbox/build$
Here I was trying to build RB for theEros Q Native, but what I choose to build doesn't seem to make a difference.
I can still use my VM should it be necessary to test/update translations, but getting it to work on a real machine instead and getting rid of the VM would be far better.
So hope there's a solution for this, but be as it may, thanks for reading and have a nice day. 🙂