gnoboot

10.4 and BETA

Fixed! LIO+noop vt-d - Raid0 block level (2 SSDs) ----------------------------------------------------------------------- CrystalDiskMark 3.0.3 x64 (C) 2007-2013 hiyohiyo Crystal Dew World : http://crystalmark.info/ ----------------------------------------------------------------------- * MB/s = 1,000,000 byte/s [sATA/300 = 300,000,000 byte/s] Sequential Read : 67.032 MB/s Sequential Write : 73.512 MB/s Random Read 512KB : 39.569 MB/s Random Write 512KB : 84.716 MB/s Random Read 4KB (QD=1) : 5.331 MB/s [ 1301.4 IOPS] Random Write 4KB (QD=1) : 3.458 MB/s [ 844.2 IOPS] Random Read 4KB (QD=32) : 54.260 MB/s [ 13247.0 IOPS] Random Write 4KB (QD=32) : 44.853 MB/s [ 10950.5 IOPS] Test : 1000 MB [G: 0.8% (0.1/10.0 GB)] (x5) Date : 2014/03/19 21:57:42 OS : Windows 7 Ultimate [6.1 Build 7600] (x64) [spoiler=]

The same reason why I tried XPEnology, but LIO iSCSI is not fully functional. Got it working eventually (file & block backend)! I'm working on fixing mpt2sas iSCSI issue with my boot image. Here's my test results: IET+noop vt-d - Raid0 block level (2 SSDs) ----------------------------------------------------------------------- CrystalDiskMark 3.0.3 x64 (C) 2007-2013 hiyohiyo Crystal Dew World : http://crystalmark.info/ ----------------------------------------------------------------------- * MB/s = 1,000,000 byte/s [sATA/300 = 300,000,000 byte/s] Sequential Read : 74.706 MB/s Sequential Write : 96.323 MB/s Random Read 512KB : 66.134 MB/s Random Write 512KB : 85.864 MB/s Random Read 4KB (QD=1) : 5.925 MB/s [ 1446.7 IOPS] Random Write 4KB (QD=1) : 6.490 MB/s [ 1584.4 IOPS] Random Read 4KB (QD=32) : 63.203 MB/s [ 15430.4 IOPS] Random Write 4KB (QD=32) : 89.203 MB/s [ 21778.0 IOPS] Test : 1000 MB [i: 0.3% (0.0/10.0 GB)] (x5) Date : 2014/03/18 20:09:15 OS : Windows 7 Ultimate [6.1 Build 7600] (x64) IET+cfq vt-d - Raid0 block level (2 SSDs) ----------------------------------------------------------------------- CrystalDiskMark 3.0.3 x64 (C) 2007-2013 hiyohiyo Crystal Dew World : http://crystalmark.info/ ----------------------------------------------------------------------- * MB/s = 1,000,000 byte/s [sATA/300 = 300,000,000 byte/s] Sequential Read : 77.226 MB/s Sequential Write : 95.568 MB/s Random Read 512KB : 64.591 MB/s Random Write 512KB : 85.319 MB/s Random Read 4KB (QD=1) : 5.907 MB/s [ 1442.2 IOPS] Random Write 4KB (QD=1) : 6.841 MB/s [ 1670.2 IOPS] Random Read 4KB (QD=32) : 60.185 MB/s [ 14693.6 IOPS] Random Write 4KB (QD=32) : 78.320 MB/s [ 19121.0 IOPS] Test : 1000 MB [i: 0.3% (0.0/10.0 GB)] (x5) Date : 2014/03/18 20:17:23 OS : Windows 7 Ultimate [6.1 Build 7600] (x64) LIO+noop vt-d - Raid0 file level (2 SSDs) ----------------------------------------------------------------------- CrystalDiskMark 3.0.3 x64 (C) 2007-2013 hiyohiyo Crystal Dew World : http://crystalmark.info/ ----------------------------------------------------------------------- * MB/s = 1,000,000 byte/s [sATA/300 = 300,000,000 byte/s] Sequential Read : 69.090 MB/s Sequential Write : 88.915 MB/s Random Read 512KB : 53.987 MB/s Random Write 512KB : 77.443 MB/s Random Read 4KB (QD=1) : 6.942 MB/s [ 1694.8 IOPS] Random Write 4KB (QD=1) : 2.786 MB/s [ 680.1 IOPS] Random Read 4KB (QD=32) : 48.489 MB/s [ 11838.2 IOPS] Random Write 4KB (QD=32) : 38.899 MB/s [ 9496.8 IOPS] Test : 1000 MB [G: 0.8% (0.1/10.0 GB)] (x5) Date : 2014/03/19 19:21:37 OS : Windows 7 Ultimate [6.1 Build 7600] (x64) LIO+cfq vt-d - Raid0 file level (2 SSDs) ----------------------------------------------------------------------- CrystalDiskMark 3.0.3 x64 (C) 2007-2013 hiyohiyo Crystal Dew World : http://crystalmark.info/ ----------------------------------------------------------------------- * MB/s = 1,000,000 byte/s [sATA/300 = 300,000,000 byte/s] Sequential Read : 63.148 MB/s Sequential Write : 87.762 MB/s Random Read 512KB : 67.471 MB/s Random Write 512KB : 75.709 MB/s Random Read 4KB (QD=1) : 6.395 MB/s [ 1561.4 IOPS] Random Write 4KB (QD=1) : 2.576 MB/s [ 628.9 IOPS] Random Read 4KB (QD=32) : 43.786 MB/s [ 10690.0 IOPS] Random Write 4KB (QD=32) : 39.534 MB/s [ 9652.0 IOPS] Test : 1000 MB [G: 0.8% (0.1/10.0 GB)] (x5) Date : 2014/03/19 19:30:53 OS : Windows 7 Ultimate [6.1 Build 7600] (x64)

Your video wasn't helpful, kernel panic was too fast . I'm thinking of driver conflict, please send `lspci -vvv` and `dmesg` from Linux box or Trantor's build.

Ok. Thanks for the fast reply. The issue is that I have a HP N54L and it doesn't support VT-d. This is the only thing that prevents me to run DSM in ESXi. When I tested OMW(OpenMediaVault) SMART worked fine with RDM, so hopefully gnoBoot will figure this one out. @gnoBoot: Do you think it's possible to get SMART working running gnoBoot and Synology DSM in ESXi with Raw Disk Mapping? I don't think so. I've read it here that you need to enable/disable RDM filter to make it work. I just couldn't remember which thread it was.

Which version? Was it only 10.3 causing kernel panics?

Try re-installing the app related to those daemons. Simply ignore it if you're not using sound devices. It's related to DSM that I can't fix:(

linux backports (compat-wireless) doesn't compile, and latest tg3 drivers requires a patch in makeflags.sh (comment UAPI).

Python 3.x is not backward compatible, have you tried synocommunity spks?

dmesg or screenshot would be helpful

Hello! first let me thank you for your hard work, i just read about gnoboot and its looking really good I tried to install it under my hyperv 2012 r2 and got your alpha 10.3 and im having the same issue like offensive1985, there is no network connection when i use the normal network adapter. But when i use the legacy one i can find the diskstation in the dsm assistant. Your advise was the one in the quote, but i have to admit i dont really understand what you mean by that. But in the meantime new versions arrived so is it possible now to run the 10.3 under hyperv with normal network or do i have to run the legacy adapter ? Of course i would also be very happy if somebody else has an advice for me ! thank you in advance greetings simmboo I don't use Hyper-V, so it's best you try it first and then share your results. It should work, though, not the latest VMware drivers. You don't need to, but 10.3 has a broken /etc/rc.network. You have to manually comment out the line that modifies MAC address. Try blacklisting e1000 using rmmod GRUB option.

Haven't tested but it should work, check this post to load the correct driver.

Provide a screenshoot. Can you check if it's similar to this bug report?

You can use insmod GRUB option to load missing drivers;) Full details posted in my blog. If it's not too much to ask, click the ADs or Bitcoin donate.

got a new mirror, check my downloads page. @eleka, was it supported before? can you post more details like `dmesg` or screenshot?

Lol, you're using the wrong zImage. That's five-zero kernel which doesn't have the XEN disk patch.

This documentation seems to be empty. Do I need a password or something? Something is fishy.... Try google cache or internet archive.

It's been available since day one. Don't forget to check the build date. Click the ADs or email me for donations:) Enjoy!

This patch works well with dsgpl-4.3 source code. It also works with dsgpl-5.0 source code but you have to manually edit the relevant files. Go start build your own kernel and if ever you give back to this community. You can't please everyone, some of them are blood sucking leeches! --- drivers/ata/libata-core.c.orig 2014-01-06 02:42:49.946152467 +0800 +++ drivers/ata/libata-core.c 2014-01-06 02:51:38.822224108 +0800 @@ -7359,7 +7359,7 @@ EXPORT_SYMBOL(funcSYNOSendDiskPortDisEvent); #endif /* MY_ABC_HERE */ -#ifdef MY_DEF_HERE +#if (defined(MY_DEF_HERE) || defined(XPENOLOGY)) int (*funcSYNOSendEboxRefreshEvent)(int portIndex) = NULL; EXPORT_SYMBOL(funcSYNOSendEboxRefreshEvent); #endif --- include/linux/syno_user.h.orig 2014-01-06 02:42:49.948152589 +0800 +++ include/linux/syno_user.h 2014-01-06 03:01:35.815989363 +0800 @@ -4,6 +4,11 @@ #define __SYNO_USER_H_ /** + * XPEnology project + */ +#define XPENOLOGY + +/** * Dsc: Synology multimedia server feature. For indexing video, photo, * music. * Ref: libsynosdk, lnxnetatalk, lnxsdk, rsync, samba, smbftpd --- kernel/sysctl.c.orig 2014-01-06 02:42:49.955153014 +0800 +++ kernel/sysctl.c 2014-01-06 02:56:38.044634463 +0800 @@ -228,7 +228,7 @@ int (*funcSYNOGetHwCapability)(CAPABILITY *) = NULL; EXPORT_SYMBOL(funcSYNOGetHwCapability); -#ifdef MY_DEF_HERE +#if (defined(MY_DEF_HERE) || defined(XPENOLOGY)) EUNIT_PWRON_TYPE (*funcSynoEunitPowerctlType)(void) = NULL; EXPORT_SYMBOL(funcSynoEunitPowerctlType); #endif --- drivers/ata/sata_mv.c.orig 2013-08-24 17:37:08.000000000 +0800 +++ drivers/ata/sata_mv.c 2014-01-07 17:11:57.921748538 +0800 @@ -3692,6 +3692,7 @@ /*FIXME - Too brutal and directly, should separate into levels*/ void syno_sata_mv_gpio_write(u8 blFaulty, const unsigned short hostnum) { +#ifndef XPENOLOGY struct Scsi_Host *shost = scsi_host_lookup(hostnum); struct ata_port *ap = NULL; void __iomem *host_mmio = NULL; @@ -3724,9 +3725,10 @@ writel(gpio_value, host_mmio + GPIO_CTL_DATA); scsi_host_put(shost); - END: +#endif /* XPENOLOGY */ return; + } EXPORT_SYMBOL(syno_sata_mv_gpio_write); #endif --- drivers/pci/quirks.c.orig 2014-02-16 07:28:49.702724291 -0800 +++ drivers/pci/quirks.c 2014-02-16 07:30:58.599035923 -0800 @@ -618,7 +618,9 @@ ( pin < 0 || pin >= 96 ) || NULL == pValue ) { +#ifndef XPENOLOGY printk("parameter error. gpiobase=%08X, pin=%d, pValue=%p\n", gpiobase, pin, pValue); +#endif /* XPENOLOGY */ goto END; } --- fs/compat_ioctl.c.orig 2014-02-16 07:32:49.409608822 -0800 +++ fs/compat_ioctl.c 2014-02-16 07:34:43.256789001 -0800 @@ -1760,12 +1760,14 @@ sprintf(buf,"'%c'", (cmd>>_IOC_TYPESHIFT) & _IOC_TYPEMASK); if (!isprint(buf[1])) sprintf(buf, "%02x", buf[1]); +#ifndef XPENOLOGY compat_printk("ioctl32(%s:%d): Unknown cmd fd(%d) " "cmd(%08x){t:%s;sz:%u} arg(%08x) on %s\n", current->comm, current->pid, (int)fd, (unsigned int)cmd, buf, (cmd >> _IOC_SIZESHIFT) & _IOC_SIZEMASK, (unsigned int)arg, fn); +#endif if (path) free_page((unsigned long)path); --- drivers/tty/vt/vt.c.orig +++ drivers/tty/vt/vt.c @@ -2474,7 +2474,7 @@ int vt_kmsg_redirect(int new) return kmsg_con; } -#ifndef MY_ABC_HERE +#ifdef MY_ABC_HERE /* * Console on virtual terminal * @@ -2574,7 +2574,7 @@ quit: } #endif -#ifndef MY_ABC_HERE +#ifdef MY_ABC_HERE static struct tty_driver *vt_console_device(struct console *c, int *index) { *index = c->index ? c->index-1 : fg_console; @@ -2942,7 +2942,7 @@ static int __init con_init(void) console_unlock(); -#ifndef MY_ABC_HERE +#ifdef MY_ABC_HERE #ifdef CONFIG_VT_CONSOLE register_console(&vt_console_driver); #endif --- drivers/scsi/sd.c.orig 2014-02-15 16:28:48.148339650 -0800 +++ drivers/scsi/sd.c 2014-02-15 16:30:15.022585662 -0800 @@ -3006,6 +3006,11 @@ { struct scsi_device *sdp = to_scsi_device(dev); +#ifdef XPENOLOGY + printk(KERN_DEBUG "%s: disk driver '%s'", __FUNCTION__, + sdp->host->hostt->name); +#endif + // iscsi #ifdef MY_ABC_HERE if(strcmp(sdp->host->hostt->name, "iSCSI Initiator over TCP/IP") == 0){ @@ -3036,12 +3041,19 @@ if (SYNO_PORT_TYPE_SATA == sdp->host->hostt->syno_port_type) { // else treat as internal disks + printk(KERN_DEBUG "%s: SATA syno_port_type %x\n", __FUNCTION__, + sdp->host->hostt->syno_port_type); return SYNO_DISK_SATA; } // sas disks if (SYNO_PORT_TYPE_SAS == sdp->host->hostt->syno_port_type) { + printk(KERN_DEBUG "%s: SAS syno_port_type %x", __FUNCTION__, + sdp->host->hostt->syno_port_type); return SYNO_DISK_SAS; } +#ifdef XPENOLOGY + printk(KERN_DEBUG "syno_disk_type_get: Got UNKNOWN port type %x\n", sdp->host->hostt->syno_port_type); +#endif return SYNO_DISK_UNKNOWN; } #endif @@ -3210,7 +3222,7 @@ #endif // try at most 5 times while (want_idx != index && - (SYNO_DISK_SATA == sdkp->synodisktype) && iRetry < 15) { + (SYNO_DISK_SATA == sdkp->synodisktype) && iRetry < 5) { /* Sometimes raid is not release all scsi disk yet. Try to delay and reget */ printk("want_idx %d index %d. delay and reget\n", want_idx, index); @@ -3295,6 +3307,8 @@ error = sd_format_disk_name(SYNO_SATA_DEVICE_PREFIX, index, gd->disk_name, DISK_NAME_LEN); break; case SYNO_DISK_USB: + error = sd_format_disk_name(SYNO_SATA_DEVICE_PREFIX, index, gd->disk_name, DISK_NAME_LEN); + break; default: #ifdef SYNO_SAS_DISK_NAME if (1 == g_is_sas_model) { @@ -3303,6 +3317,10 @@ } #endif error = sd_format_disk_name(SYNO_SATA_DEVICE_PREFIX, index, gd->disk_name, DISK_NAME_LEN); +#ifdef XPENOLOGY + printk(KERN_DEBUG "sd_probe: Got UNKNOWN disk %s with index %x\n", gd->disk_name, index); + gd->systemDisk = 1; +#endif break; } #else

Oh well, I've even released the kernel patch. So all you F*CK!*G suckers can build your own release! Did anybody here even release their patch? Yeah Andy was the first one, but it doesn't apply with the latest dsgpl (4.3) source code. --- drivers/ata/libata-core.c.orig 2014-01-06 02:42:49.946152467 +0800 +++ drivers/ata/libata-core.c 2014-01-06 02:51:38.822224108 +0800 @@ -7359,7 +7359,7 @@ EXPORT_SYMBOL(funcSYNOSendDiskPortDisEvent); #endif /* MY_ABC_HERE */ -#ifdef MY_DEF_HERE +#if (defined(MY_DEF_HERE) || defined(XPENOLOGY)) int (*funcSYNOSendEboxRefreshEvent)(int portIndex) = NULL; EXPORT_SYMBOL(funcSYNOSendEboxRefreshEvent); #endif --- include/linux/syno_user.h.orig 2014-01-06 02:42:49.948152589 +0800 +++ include/linux/syno_user.h 2014-01-06 03:01:35.815989363 +0800 @@ -4,6 +4,11 @@ #define __SYNO_USER_H_ /** + * XPEnology project + */ +#define XPENOLOGY + +/** * Dsc: Synology multimedia server feature. For indexing video, photo, * music. * Ref: libsynosdk, lnxnetatalk, lnxsdk, rsync, samba, smbftpd --- kernel/sysctl.c.orig 2014-01-06 02:42:49.955153014 +0800 +++ kernel/sysctl.c 2014-01-06 02:56:38.044634463 +0800 @@ -228,7 +228,7 @@ int (*funcSYNOGetHwCapability)(CAPABILITY *) = NULL; EXPORT_SYMBOL(funcSYNOGetHwCapability); -#ifdef MY_DEF_HERE +#if (defined(MY_DEF_HERE) || defined(XPENOLOGY)) EUNIT_PWRON_TYPE (*funcSynoEunitPowerctlType)(void) = NULL; EXPORT_SYMBOL(funcSynoEunitPowerctlType); #endif --- drivers/ata/sata_mv.c.orig 2013-08-24 17:37:08.000000000 +0800 +++ drivers/ata/sata_mv.c 2014-01-07 17:11:57.921748538 +0800 @@ -3692,6 +3692,7 @@ /*FIXME - Too brutal and directly, should separate into levels*/ void syno_sata_mv_gpio_write(u8 blFaulty, const unsigned short hostnum) { +#ifndef XPENOLOGY struct Scsi_Host *shost = scsi_host_lookup(hostnum); struct ata_port *ap = NULL; void __iomem *host_mmio = NULL; @@ -3724,9 +3725,10 @@ writel(gpio_value, host_mmio + GPIO_CTL_DATA); scsi_host_put(shost); - END: +#endif /* XPENOLOGY */ return; + } EXPORT_SYMBOL(syno_sata_mv_gpio_write); #endif --- drivers/pci/quirks.c.orig 2014-02-16 07:28:49.702724291 -0800 +++ drivers/pci/quirks.c 2014-02-16 07:30:58.599035923 -0800 @@ -618,7 +618,9 @@ ( pin < 0 || pin >= 96 ) || NULL == pValue ) { +#ifndef XPENOLOGY printk("parameter error. gpiobase=%08X, pin=%d, pValue=%p\n", gpiobase, pin, pValue); +#endif /* XPENOLOGY */ goto END; } --- fs/compat_ioctl.c.orig 2014-02-16 07:32:49.409608822 -0800 +++ fs/compat_ioctl.c 2014-02-16 07:34:43.256789001 -0800 @@ -1760,12 +1760,14 @@ sprintf(buf,"'%c'", (cmd>>_IOC_TYPESHIFT) & _IOC_TYPEMASK); if (!isprint(buf[1])) sprintf(buf, "%02x", buf[1]); +#ifndef XPENOLOGY compat_printk("ioctl32(%s:%d): Unknown cmd fd(%d) " "cmd(%08x){t:%s;sz:%u} arg(%08x) on %s\n", current->comm, current->pid, (int)fd, (unsigned int)cmd, buf, (cmd >> _IOC_SIZESHIFT) & _IOC_SIZEMASK, (unsigned int)arg, fn); +#endif if (path) free_page((unsigned long)path); --- drivers/tty/vt/vt.c.orig +++ drivers/tty/vt/vt.c @@ -2474,7 +2474,7 @@ int vt_kmsg_redirect(int new) return kmsg_con; } -#ifndef MY_ABC_HERE +#ifdef MY_ABC_HERE /* * Console on virtual terminal * @@ -2574,7 +2574,7 @@ quit: } #endif -#ifndef MY_ABC_HERE +#ifdef MY_ABC_HERE static struct tty_driver *vt_console_device(struct console *c, int *index) { *index = c->index ? c->index-1 : fg_console; @@ -2942,7 +2942,7 @@ static int __init con_init(void) console_unlock(); -#ifndef MY_ABC_HERE +#ifdef MY_ABC_HERE #ifdef CONFIG_VT_CONSOLE register_console(&vt_console_driver); #endif --- drivers/scsi/sd.c.orig 2014-02-15 16:28:48.148339650 -0800 +++ drivers/scsi/sd.c 2014-02-15 16:30:15.022585662 -0800 @@ -3006,6 +3006,11 @@ { struct scsi_device *sdp = to_scsi_device(dev); +#ifdef XPENOLOGY + printk(KERN_DEBUG "%s: disk driver '%s'", __FUNCTION__, + sdp->host->hostt->name); +#endif + // iscsi #ifdef MY_ABC_HERE if(strcmp(sdp->host->hostt->name, "iSCSI Initiator over TCP/IP") == 0){ @@ -3036,12 +3041,19 @@ if (SYNO_PORT_TYPE_SATA == sdp->host->hostt->syno_port_type) { // else treat as internal disks + printk(KERN_DEBUG "%s: SATA syno_port_type %x\n", __FUNCTION__, + sdp->host->hostt->syno_port_type); return SYNO_DISK_SATA; } // sas disks if (SYNO_PORT_TYPE_SAS == sdp->host->hostt->syno_port_type) { + printk(KERN_DEBUG "%s: SAS syno_port_type %x", __FUNCTION__, + sdp->host->hostt->syno_port_type); return SYNO_DISK_SAS; } +#ifdef XPENOLOGY + printk(KERN_DEBUG "syno_disk_type_get: Got UNKNOWN port type %x\n", sdp->host->hostt->syno_port_type); +#endif return SYNO_DISK_UNKNOWN; } #endif @@ -3210,7 +3222,7 @@ #endif // try at most 5 times while (want_idx != index && - (SYNO_DISK_SATA == sdkp->synodisktype) && iRetry < 15) { + (SYNO_DISK_SATA == sdkp->synodisktype) && iRetry < 5) { /* Sometimes raid is not release all scsi disk yet. Try to delay and reget */ printk("want_idx %d index %d. delay and reget\n", want_idx, index); @@ -3295,6 +3307,8 @@ error = sd_format_disk_name(SYNO_SATA_DEVICE_PREFIX, index, gd->disk_name, DISK_NAME_LEN); break; case SYNO_DISK_USB: + error = sd_format_disk_name(SYNO_SATA_DEVICE_PREFIX, index, gd->disk_name, DISK_NAME_LEN); + break; default: #ifdef SYNO_SAS_DISK_NAME if (1 == g_is_sas_model) { @@ -3303,6 +3317,10 @@ } #endif error = sd_format_disk_name(SYNO_SATA_DEVICE_PREFIX, index, gd->disk_name, DISK_NAME_LEN); +#ifdef XPENOLOGY + printk(KERN_DEBUG "sd_probe: Got UNKNOWN disk %s with index %x\n", gd->disk_name, index); + gd->systemDisk = 1; +#endif break; } #else

vfat-password [spoiler=]gn0Boo7-aLpha10 If you can wait, next release will be this weekend. And I will try to get all request included, specially XEN support and 5.0-4458.

It's already supported on alpha10.2 (i.e. /proc/acpi/event), see changelog at my blog:) Well, just always check my blog to see what it is. I missed the new tigon driver somehow during my build. So workaround is `insmod /lib/modules/tg3_new.ko` in /etc/rc.local, or add 'insmod=tg3' GRUB option if you are installing to a new machine.

Something big is coming in the next few days... [spoiler=]Click the ADs on my blog or email me for donation. Enjoy!

Try alpha8 that has multimedia support enabled.

[spoiler=]