<div dir="ltr"><div>You're right, this is definitely what's missing:</div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><span style="font-family:monospace"><span style="color:rgb(0,0,0);background-color:rgb(255,255,255)"> linux-vdso.so.1 (0x00007ffeb2163000)
</span><br> libfuse.so.2 => /lib/x86_64-linux-gnu/libfuse.so.2 (0x00007ffb2d6b7000)
<br> libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007ffb2d694000)
<br> libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007ffb2d4a3000)
<br> libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007ffb2d49d000)
<br> /lib64/ld-linux-x86-64.so.2 (0x00007ffb2d728000)<br></span></div></blockquote><div><br></div><div>And actually when running configure, I notice this:</div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><span style="font-family:monospace"><span style="color:rgb(0,0,0);background-color:rgb(255,255,255)">checking lz4.h usability... no
</span><br>checking lz4.h presence... no
<br>checking for lz4.h... no<br></span></div></blockquote><div><br></div><div>Not sure what I'm missing here though...</div><div><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><span style="font-family:monospace"><span style="color:rgb(0,0,0);background-color:rgb(255,255,255)">$ apt list --installed | grep lz4
</span><br>lib<span style="font-weight:bold;color:rgb(255,84,84);background-color:rgb(255,255,255)">lz4</span><span style="color:rgb(0,0,0);background-color:rgb(255,255,255)">-1/now 1.9.3-2 amd64 [installed,local]
</span><br><span style="font-weight:bold;color:rgb(255,84,84);background-color:rgb(255,255,255)">lz4</span><span style="color:rgb(0,0,0);background-color:rgb(255,255,255)">/now 1.9.3-2 amd64 [installed,local]</span><br></span></div></blockquote></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, 20 Aug 2021 at 16:27, Gao Xiang <<a href="mailto:xiang@kernel.org">xiang@kernel.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On Fri, Aug 20, 2021 at 04:16:20PM +0300, Igor Eisberg wrote:<br>
> You're quicker than expected, thanks for answering.<br>
> Not sure how to check if lz4 was builtin, but considering that erofsfuse is<br>
> only about 34.5KB (stripped) I would guess not?<br>
> Here's the output of erofsfuse -d (it prints this but never exists back to<br>
> shell unless I do Ctrl+C):<br>
<br>
Yeah, it will run erofsfuse in foreground, and you need to access the<br>
erofs compressed files, and then check the printed result.<br>
<br>
But I think there is a easier way to check if lz4 was linked, just type<br>
ldd <your erofsfuse program><br>
<br>
if lz4 was linked, it will print some message like below:<br>
linux-vdso.so.1 (0x00007ffee176e000)<br>
libfuse.so.2 => /lib/x86_64-linux-gnu/libfuse.so.2 (0x00007f8e21f24000)<br>
liblz4.so.1 => /lib/x86_64-linux-gnu/liblz4.so.1 (0x00007f8e21f01000)<br>
libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f8e21ee0000)<br>
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f8e21d1f000)<br>
libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007f8e21d1a000)<br>
/lib64/ld-linux-x86-64.so.2 (0x00007f8e21f91000)<br>
<br>
Thanks,<br>
Gao Xiang<br>
<br>
> <br>
> erofsfuse 1.3<br>
> ><br>
> > disk: product.img<br>
> ><br>
> > mountpoint: product-mnt<br>
> ><br>
> > dbglevel: 7<br>
> ><br>
> > FUSE library version: 2.9.9<br>
> ><br>
> > nullpath_ok: 0<br>
> ><br>
> > nopath: 0<br>
> ><br>
> > utime_omit_ok: 0<br>
> ><br>
> > unique: 1, opcode: INIT (26), nodeid: 0, insize: 56, pid: 0<br>
> ><br>
> > INIT: 7.27<br>
> ><br>
> > flags=0x003ffffb<br>
> ><br>
> > max_readahead=0x00020000<br>
> ><br>
> > EROFS: erofsfuse_init() Line[23] Using FUSE protocol 7.27<br>
> ><br>
> > INIT: 7.19<br>
> ><br>
> > flags=0x00000011<br>
> ><br>
> > max_readahead=0x00020000<br>
> ><br>
> > max_write=0x00020000<br>
> ><br>
> > max_background=0<br>
> ><br>
> > congestion_threshold=0<br>
> ><br>
> > unique: 1, success, outsize: 40<br>
> ><br>
> ><br>
> On Fri, 20 Aug 2021 at 15:49, Gao Xiang <<a href="mailto:xiang@kernel.org" target="_blank">xiang@kernel.org</a>> wrote:<br>
> <br>
> > Hi Igor,<br>
> ><br>
> > On Fri, Aug 20, 2021 at 03:34:05PM +0300, Igor Eisberg wrote:<br>
> > > Hey there, getting straight to the point.<br>
> > > Our team is using Debian 10, in which erofs mounting is not supported and<br>
> > > we have no option of updating the kernel, nor do we have sudo permissions<br>
> > > on this server.<br>
> > ><br>
> > > Our only choice is to use erofsfuse to mount an Android image<br>
> > (compression<br>
> > > was used on that image), for the sole purpose of extracting its contents<br>
> > to<br>
> > > another folder for processing.<br>
> > > Tried on Debian 10, pop_OS! and even the latest Kubuntu (where native<br>
> > > mounting is supported), but on all of them I could not copy files which<br>
> > are<br>
> > > compressed from the mounted image to another location (ext4 file system).<br>
> > ><br>
> > > The error I'm getting is: "Operation not supported (95)"<br>
> > ><br>
> ><br>
> > Thanks for your feedback.<br>
> ><br>
> > Could you check if lz4 was built-in when building erofsfuse? I guess<br>
> > that is the reason (lack of lz4 support builtin).<br>
> ><br>
> > If not, could you add -d to erofsfuse when starting up?<br>
> ><br>
> > Thanks,<br>
> > Gao Xiang<br>
> ><br>
> > > Notes:<br>
> > > * Only extremely small (< 1 KB) files which are stored uncompressed are<br>
> > > copied successfully.<br>
> > > * Copying works perfectly when mounting the image with "sudo mount" on<br>
> > the<br>
> > > latest Kubuntu, so it has to be something with erofsfuse.<br>
> > ><br>
> > > Anything you can do to help resolve this?<br>
> > ><br>
> > > Best,<br>
> > > Igor.<br>
> ><br>
</blockquote></div>