Everyday I’m using a tower PC running Ubuntu 18.04 to take care of this blog, but when I travel it’s obviously not so convenient, so a few years ago I bought an Acer Aspire E5-421G laptop powered by an AMD A4-6210 processor with 4GB RAM, 512GB HDD, and a 14″ display. I installed Ubuntu on the laptop and it works, but with 4GB RAM, it’s not always usable while multitasking. For example I can run Thunderbird and Firefox, but if I ever make a Skype call for example, the system becomes unusable, and I have to close one of the programs. Tasks like video editing are also quite slow on the machine.
So since I’m going to travel in a few weeks, I decided I needed a new laptop. My requirements were 8GB RAM, SSD and HDD support, a 15″ display, the ability to run Ubuntu 18.04, and possibly a processor with a performance close to the AMD FX8350 processor in my tower. I also had a target budget of around 20,000 Baht (~$600 US). After doing some research online, I found yet again an Acer laptop that met all my requirements: Acer Aspire 3 A315-41G (-R468) with an AMD Ryzen 7 2700U quad core/octa thread processor (15W TDP), 8GB DDR4, 1TB HDD, and a 15.6″ Full HD display (Amazon Link). Where’s the M.2 SSD? According to a YouTube video on the right of the RAM compartment.
So I went to my local shop, and I could find that exact model for 20,990 THB (~$632). All good until I asked the seller to confirm there was indeed an M.2 SSD socket on the motherboard, and he answered something in the line of “nope, only for Intel laptops, not for AMD laptops”. After showing him the YouTube video, he and his colleague decided to look through the RAM compartment to try to see, and it really look like there was one… So finally, I went ahead and purchased the thing…
I got a 10-in-1 gift kit with some junk, a USB mouse, and a Predator carrying bag. When we start the laptop we can see it’s running Linpus Linux (Lite) just like in my previous Acer laptop, which made me confident Ubuntu would just run out of the box. Linpus Linux Lite is a sort of “FreeDOS equivalent for Linux” as it’s a command line only OS that Acer uses to sell laptops without Windows.
We can double check the specs by accessing the BIOS while pressing the F2 at boot time.
InsydeH20 setup utility shows an AMD Ryzen 7 2700U with Radeon Vega Mobile Gfx, a TOSHIBA hard drive, and 8192 MB physical memory.
The BIOS does not have an awful lot of options however.
But let’s go have a look at the laptop itself first. The left side comes with a Kesington lock, an Ethernet port, an HDMI port, a USB 3.0 port, and an SD slot.
The right side include a 3.5mm headphone jack, two USB ports, some LEDs, and the power jack.
The bottom of the laptop is more interesting. We have a 2.5″ SATA drive bay, a RAM compartment, and what should be a socket for M.2 2280 SATA SSD, but not easily accessible. The video linked in the introduction mentions to remove all screws to take out the bottom cover. There’s just one little problem: I’m in Thailand, and we don’t have laws providing “warranty void stickers” like the FCC does in the US, so if I ever damage it, I’d lose the 2-year warranty that comes with my laptop.
We can find the 1TB hard drive and two memory sticks once we take out the user serviceable covers. The M.2 socket is on the right, and the warranty void sticker a bit on the left, so I took out of screws on the right of the sticker and a few on its left, and try to see if I could install an M.2 SSD that way.
That’s the most I can safely lift the cover, and we can indeed confirm a M.2 socket is on the motherboard. But it’s not really easy to insert the SSD, and the final step (adding the screws) would be even more challenging. So I’ve decided to use the laptop without SSD for now, and will only install the M.2 SSD and likely void the warranty if I become frustrated with performance. Anyway this is madness to make a motherboard with an M.2 SSD socket and not easily expose it to customers!
Time to install Ubuntu 18.04.1. After downloading the ISO I flash a (Kingston DataTraveler) USB flash drive with USB Disk Creator, inserted it into the USB 3.0 port of the laptop, access the BIOS and it was detected properly as a boot device, even showing in first place.
So I exited the BIOS, and within a few seconds I was in the menu asking me to try or install Ubuntu. I selected “Install Ubuntu”, but the screen stayed black a little to long, before spewing some “ACPI Error”, “Firmware Bug”, and several messages like “BUG: soft lockup – CPU#0 stuck for 22s!”
It does not look to good. First I tried to move the USB drive to a USB 2.0 port. No luck, and then I went to Advanced settings in the BIOS to disable AMD-SVM and AMD-IOMMU since I could see some message related to the later.
But it did not work either. But the way, the SATA configuration above shows both SATA0 and SATA1 interface so I’d expect both the 2.5″ SATA and M.2 SATA to work.
I then noticed the BIOS was a bit old (2017), so I went to the laptop support page to look for some updates, and indeed BIOS 1.08 (May 2018) is now available while my laptop is running 1.03.
So I downloaded it to find out it was an EXE file meant to run in Windows. Right… After searching some info in the web, more frustration came about because Acer just tells you to install Windows to upgrade the BIOS. More madness from Acer. I was stuck, so no choice, I prepared a Windows 10 boot CD, installed Windows, downloaded the BIOS installer, and ran it.
It will reboot the system to perform the update, and everything went smoothly with BIOS v1.08 now installed in the Laptop.
Good, let’s wipe out Windows 10 from the USB drive, and reflash Ubuntu 18.04 ISO. Sadly back to square 1 as the ACPI errors were back.
I noticed some other people had troubles with AMD Ryzen processor and they used a more recent Linux 4.17. So I tried to respin Ubuntu 18.04 with a mainline kernel, but somehow I never managed to make the image boot at all. It’s not detected as a bootable image in the BIOS.
Some more research led me to two bug reports here and there where people have similar issues with AMD Ryzen 5/7 based Acer laptops. It looks like it may be a BIOS issue, but people tried various kernel parameters to work around the issue.
What worked for me while in the installer was to press “e” with “Install Ubuntu” highlighted in order to edit the parameter, and I modified the linux line by adding “pci=noacpi” at the end. My first attempt with “acpi=off” as shown above did not work. I then pressed F10 to boot with my modification and could install Ubuntu without issues. The touchpad did not work, but I found out later that I had to press F7 to enable the touchpad.
I tested all main features of the laptop:
- 1920×1080 display – OK
- HDMI output – OK (extended display works)
- USB 2.0 ports – OK
- USB 3.0 port – OK (Tested with USB 3.0 HDD at ~100MB/s)
- Gigabit Ethernet and 802.11ac Wi-Fi – OK
- Bluetooth – OK (Tested with Bluetooth headphones)
- Keyboard – OK
- Touchpad – OK (tested “basic” mode only)
- Webcam, speakers, and audio jack – OK
- SD Card – Failed [Update August 14, 2019: OK with Ubuntu 18.04.3 and Linux 5.0.0]
That’s the error messages I get when inserting an SD card:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 |
[22614.698950] mmc0: cannot verify signal voltage switch [22614.699457] mmc0: Skipping voltage switch [22615.321724] mmc0: error -110 whilst initialising SD card [22714.978965] mmc0: cannot verify signal voltage switch [22715.417849] mmc0: error -110 whilst initialising SD card [22732.345118] mmc0: card never left busy state [22732.345126] mmc0: error -110 whilst initialising SD card [22732.760465] rtsx_pci_sdmmc rtsx_pci_sdmmc.0: card claims to support voltages below defined range [22732.760468] rtsx_pci_sdmmc rtsx_pci_sdmmc.0: no support for card's volts [22732.760470] mmc0: error -22 whilst initialising SDIO card [22735.029015] mmc0: card never left busy state [22735.029024] mmc0: error -110 whilst initialising SD card [22736.050138] mmc0: cannot verify signal voltage switch [22736.484982] mmc0: error -110 whilst initialising SD card |
I’ve not checked battery life so far, and that’s one of the potential issue with disabling ACPI. Battery life is not really that important to me since I’ll work in rooms when I have WiFi (and a power socket). As a pointer, after a full charge, Ubuntu indicates 2 hours and 7 minutes remaining when disconnecting the power supply. If that’s the case it’s pretty poor battery life. I’ll confirm in the comments section once I do some more testing with the battery.
Also note that screen brightness is set to the lowest level at boot time, and the touchpad is disabled after boot/reboot which means I have to press F7 whenever I need t use the touchpad.
Some system info for people interested in details:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 |
jaufranc@cnx-laptop-4:~$ inxi -Fc0 System: Host: cnx-laptop-4 Kernel: 4.15.0-32-generic x86_64 bits: 64 Console: tty 1 Distro: Ubuntu 18.04.1 LTS Machine: Device: laptop System: Acer product: Aspire A315-41G v: V1.08 serial: N/A Mobo: RR model: Metapod_RR v: V1.08 serial: N/A UEFI: Insyde v: V1.08 date: 05/23/2018 Battery BAT1: charge: 34.6 Wh 100.0% condition: 34.6/36.7 Wh (94%) CPU: Quad core AMD Ryzen 7 2700U with Radeon Vega Mobile Gfx (-MT-MCP-) cache: 2048 KB clock speeds: max: 2200 MHz 1: 1969 MHz 2: 2294 MHz 3: 2211 MHz 4: 1471 MHz 5: 1623 MHz 6: 1561 MHz 7: 1573 MHz 8: 1535 MHz Graphics: Card-1: Advanced Micro Devices [AMD/ATI] Topaz XT [Radeon R7 M260/M265 / M340/M360 / M440/M445] Card-2: Advanced Micro Devices [AMD/ATI] Vega [Radeon Vega 8 Mobile] Display Server: N/A drivers: amdgpu,amdgpu tty size: 80x24 Advanced Data: N/A out of X Audio: Card-1 Advanced Micro Devices [AMD] Device 15e3 driver: snd_hda_intel Card-2 Advanced Micro Devices [AMD/ATI] Device 15de driver: snd_hda_intel Sound: Advanced Linux Sound Architecture v: k4.15.0-32-generic Network: Card-1: Realtek RTL8111/8168/8411 PCIE Gigabit Ethernet Controller driver: r8169 IF: enp2s0f1 state: up speed: 1000 Mbps duplex: full mac: 98:28:a6:0f:06:07 Card-2: Qualcomm Atheros QCA9377 802.11ac Wireless Network Adapter driver: ath10k_pci IF: wlp3s0 state: up mac: 70:c9:4e:b7:84:77 Drives: HDD Total Size: 2000.4GB (29.4% used) ID-1: /dev/sda model: TOSHIBA_MQ04ABF1 size: 1000.2GB ID-2: USB /dev/sdb model: Expansion size: 1000.2GB Partition: ID-1: / size: 916G used: 12G (2%) fs: ext4 dev: /dev/sda2 RAID: No RAID devices: /proc/mdstat, md_mod kernel module present Sensors: System Temperatures: cpu: 59.4C mobo: N/A gpu: 50.0,0.0 Fan Speeds (in rpm): cpu: N/A Info: Processes: 365 Uptime: 6:23 Memory: 3276.9/6931.1MB Init: systemd runlevel: 5 Client: Shell (bash) inxi: 2.3.56 |
It should also be noted I can see several warning related to the DRM (GPU) driver in dmesg, but I did not notice any side effect from a user perspective:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 |
[20251.984207] [drm:generic_reg_wait [amdgpu]] *ERROR* REG_WAIT timeout 1us * 100 tries - tgn10_lock line:566 [20251.984319] WARNING: CPU: 4 PID: 1522 at /build/linux-wuhukg/linux-4.15.0/drivers/gpu/drm/amd/amdgpu/../display/dc/dc_helper.c:190 generic_reg_wait+0xe8/0x120 [amdgpu] [20251.984320] Modules linked in: binfmt_misc rfcomm ccm cmac bnep nls_iso8859_1 arc4 snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec_hdmi snd_hda_intel snd_hda_codec uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_v4l2 videobuf2_core snd_hda_core edac_mce_amd btusb snd_hwdep ath10k_pci ath10k_core videodev kvm_amd media snd_pcm btrtl ath btbcm btintel snd_seq_midi snd_seq_midi_event kvm acer_wmi mac80211 bluetooth snd_rawmidi irqbypass crct10dif_pclmul snd_seq snd_seq_device snd_timer crc32_pclmul cfg80211 sparse_keymap rtsx_pci_ms ghash_clmulni_intel joydev input_leds snd pcbc ecdh_generic memstick serio_raw aesni_intel soundcore k10temp shpchp mac_hid aes_x86_64 crypto_simd glue_helper cryptd tpm_crb wmi_bmof sch_fq_codel parport_pc ppdev lp parport ip_tables x_tables autofs4 amdkfd [20251.984370] amd_iommu_v2 amdgpu rtsx_pci_sdmmc chash i2c_algo_bit ttm drm_kms_helper psmouse ahci syscopyarea r8169 rtsx_pci sysfillrect sysimgblt mii fb_sys_fops libahci i2c_piix4 drm video wmi [20251.984390] CPU: 4 PID: 1522 Comm: Xorg Tainted: G W 4.15.0-32-generic #35-Ubuntu [20251.984392] Hardware name: Acer Aspire A315-41G/Metapod_RR, BIOS V1.08 05/23/2018 [20251.984456] RIP: 0010:generic_reg_wait+0xe8/0x120 [amdgpu] [20251.984458] RSP: 0018:ffffb4be427cb7d8 EFLAGS: 00010297 [20251.984461] RAX: 0000000000000001 RBX: 0000000000000065 RCX: 0000000000000000 [20251.984463] RDX: 0000000000000000 RSI: ffff980a0ed16498 RDI: ffff980a0ed16498 [20251.984464] RBP: ffffb4be427cb818 R08: 0000000000000000 R09: 0000000000000752 [20251.984466] R10: 0000000000000002 R11: ffffffffad55380d R12: 0000000000000001 [20251.984467] R13: ffff980a06756080 R14: 0000000000000100 R15: 0000000000000001 [20251.984470] FS: 00007f23be766600(0000) GS:ffff980a0ed00000(0000) knlGS:0000000000000000 [20251.984472] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [20251.984473] CR2: 00007fadb22d6000 CR3: 0000000242270000 CR4: 00000000003406e0 [20251.984475] Call Trace: [20251.984548] tgn10_lock+0xa2/0xb0 [amdgpu] [20251.984614] program_all_pipe_in_tree+0x804/0x8b0 [amdgpu] [20251.984620] ? free_one_page+0x76/0x400 [20251.984684] ? amdgpu_cgs_read_register+0x14/0x20 [amdgpu] [20251.984748] dcn10_apply_ctx_for_surface+0x498/0x4f0 [amdgpu] [20251.984810] dc_commit_state+0x2aa/0x500 [amdgpu] [20251.984880] amdgpu_dm_atomic_commit_tail+0x2cd/0xa50 [amdgpu] [20251.984926] ? amdgpu_bo_pin_restricted+0x1b5/0x2a0 [amdgpu] [20251.984932] ? _cond_resched+0x19/0x40 [20251.984934] ? wait_for_completion_interruptible+0x35/0x180 [20251.984946] commit_tail+0x42/0x70 [drm_kms_helper] [20251.984955] drm_atomic_helper_commit+0x10c/0x120 [drm_kms_helper] [20251.985020] amdgpu_dm_atomic_commit+0x87/0xa0 [amdgpu] [20251.985041] drm_atomic_commit+0x51/0x60 [drm] [20251.985050] drm_atomic_helper_set_config+0x7c/0x90 [drm_kms_helper] [20251.985067] __drm_mode_set_config_internal+0x6b/0x120 [drm] [20251.985083] drm_mode_setcrtc+0x47f/0x660 [drm] [20251.985100] ? drm_mode_getcrtc+0x190/0x190 [drm] [20251.985114] drm_ioctl_kernel+0x5f/0xb0 [drm] [20251.985128] drm_ioctl+0x31b/0x3d0 [drm] [20251.985143] ? drm_mode_getcrtc+0x190/0x190 [drm] [20251.985187] amdgpu_drm_ioctl+0x4f/0x90 [amdgpu] [20251.985194] do_vfs_ioctl+0xa8/0x630 [20251.985197] ? vfs_read+0x115/0x130 [20251.985201] SyS_ioctl+0x79/0x90 [20251.985205] do_syscall_64+0x73/0x130 [20251.985209] entry_SYSCALL_64_after_hwframe+0x3d/0xa2 [20251.985211] RIP: 0033:0x7f23bbb5b5d7 [20251.985213] RSP: 002b:00007ffe9f3414b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [20251.985216] RAX: ffffffffffffffda RBX: 00007ffe9f3414f0 RCX: 00007f23bbb5b5d7 [20251.985217] RDX: 00007ffe9f3414f0 RSI: 00000000c06864a2 RDI: 000000000000000e [20251.985218] RBP: 00007ffe9f3414f0 R08: 0000000000000000 R09: 000056019dbefee0 [20251.985220] R10: 00007ffe9f3415b0 R11: 0000000000000246 R12: 00000000c06864a2 [20251.985221] R13: 000000000000000e R14: 0000000000000000 R15: 000056019dbefee0 [20251.985223] Code: 31 f6 44 8b 45 10 44 89 e1 48 c7 c7 64 38 4e c0 89 45 d4 52 48 c7 c2 a8 be 4d c0 e8 e3 5c d7 ff 41 83 7d 20 01 58 8b 45 d4 74 02 <0f> 0b 48 8d 65 d8 5b 41 5c 41 5d 41 5e 41 5f 5d c3 c7 45 c4 23 [20251.985271] ---[ end trace 7621a5a490da2216 ]--- [20251.985434] [drm] DC: Cursor address is 0! [20251.986493] [drm] DC: Cursor address is 0! [20251.987410] [drm] {1920x1080, 2250x1132@152840Khz} [20251.996250] [drm] HBRx2 pass VS=1, PE=0 |
Finally, I’ve run Octane 2.0 in Firefox to compare the performance between my AMD FX8350 based PC, and Ryzen 7 based Acer Aspire 3 A315-41G laptop.
The laptop beats my PC in every benchmarks except one. AMD Ryzen 7 2700U is expected to have better single core performance, but since it’s has 4C/8T instead of 8C/8T on AMD FX8350 highly parallel workload may run slightly faster on the older CPU.
It was a struggle to make it work, but so far I’m relatively happy with the results, even if everything is not perfect, i.e. I have not installed an M.2 SSD (yet), and the SD card reader does not seem to work just yet. Battery life may also be a concern at least until the ACPI issue is resolved.
Jean-Luc started CNX Software in 2010 as a part-time endeavor, before quitting his job as a software engineering manager, and starting to write daily news, and reviews full time later in 2011.
Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress
