| Message ID | 2c2eef3c9a2f57e5609100a4864715ccf253d30f.1631713483.git.geert+renesas@glider.be (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | sh: pgtable-3level: Fix cast to pointer from integer of different size | expand |
Hi Geert, On Wed, 15 Sept 2021 at 22:52, Geert Uytterhoeven <geert+renesas@glider.be> wrote: > Compile-tested only, as the remote Migo-R is MIA. > Anyone with e.g. a Migo-R or Ecovec to boot-test? Yes. It's 11pm here right now so I'll do it tomorrow and get back to you. Cheers, Daniel
On Wed, 15 Sept 2021 at 22:52, Geert Uytterhoeven <geert+renesas@glider.be> wrote: > Anyone with e.g. a Migo-R or Ecovec to boot-test? Tested-by: Daniel Palmer <daniel@thingy.jp> --- Massive boot log from an ecovec --- Bytes transferred = 25788416 (1898000 hex) ## Booting kernel from Legacy Image at 89000000 ... Image Name: Linux-5.15.0-rc1 Image Type: SuperH Linux Kernel Image (gzip compressed) Data Size: 3175191 Bytes = 3 MB Load Address: 88001000 Entry Point: 88002000 Verifying Checksum ... OK Uncompressing Kernel Image ... OK Linux version 5.15.0-rc1 (daniel@shiro) (sh4-buildroot-linux-gnu-gcc.br_real (Buildroot 2021.02-715-g5e3799213236) 10.3.0, GNU ld (GNU Binutils) 2.36.1) #6 Thu Sep 16 21 Boot params: ... MOUNT_ROOT_RDONLY - 00000000 ... RAMDISK_FLAGS - 00000000 ... ORIG_ROOT_DEV - 00000000 ... LOADER_TYPE - 00000000 ... INITRD_START - 00000000 ... INITRD_SIZE - 00000000 Memory limited to 248MB Booting machvec: R0P7724 (EcoVec) initrd disabled Zone ranges: Normal [mem 0x0000000008000000-0x00000000177fffff] Movable zone start for each node Early memory node ranges node 0: [mem 0x0000000008000000-0x00000000177fffff] Initmem setup node 0 [mem 0x0000000008000000-0x00000000177fffff] Built 1 zonelists, mobility grouping on. Total pages: 62930 Kernel command line: console=tty0, console=ttySC0,115200 phram.phram=rootfs,0x8a000000,32Mi memmap=32M$0x8a000000 root=/dev/mmcblk0 ro rootwait mem=248M memchunk.vpu=8mm Unknown command line parameters: memmap=32M$0x8a000000 Dentry cache hash table entries: 32768 (order: 5, 131072 bytes, linear) Inode-cache hash table entries: 16384 (order: 4, 65536 bytes, linear) Sorting __ex_table... mem auto-init: stack:off, heap alloc:off, heap free:off PVR=10300b00 CVR=7144040d PRR=00002200 I-cache : n_ways=4 n_sets=256 way_incr=8192 I-cache : entry_mask=0x00001fe0 alias_mask=0x00001000 n_aliases=2 D-cache : n_ways=4 n_sets=256 way_incr=8192 D-cache : entry_mask=0x00001fe0 alias_mask=0x00001000 n_aliases=2 S-cache : n_ways=4 n_sets=2048 way_incr=65536 S-cache : entry_mask=0x0000ffe0 alias_mask=0x0000f000 n_aliases=16 virtual kernel memory layout: fixmap : 0xdffd7000 - 0xdffff000 ( 160 kB) vmalloc : 0xc0000000 - 0xdffd5000 ( 511 MB) lowmem : 0x88000000 - 0x97800000 ( 248 MB) (cached) : 0xa0000000 - 0xc0000000 ( 512 MB) (uncached) .init : 0x884eb000 - 0x8850a000 ( 124 kB) .data : 0x883ab8e8 - 0x884ea3ec (1274 kB) .text : 0x88001000 - 0x883ab8e8 (3754 kB) Memory: 246224K/253952K available (3750K kernel code, 266K rwdata, 1004K rodata, 124K init, 89K bss, 7724K reserved, 0K cma-reserved) SLUB: HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1 NR_IRQS: 8, nr_irqs: 8, preallocated irqs: 0 intc: Registered controller 'sh7724' with 83 IRQs random: get_random_bytes called from start_kernel+0x3d6/0x740 with crng_init=0 Console: colour dummy device 80x25 printk: console [tty0] enabled sh-cmt-32.0: ch0: used for clock events sh-cmt-32.0: ch0: used for periodic clock events sh-cmt-32.0: ch0: used as clock source clocksource: sh-cmt-32.0: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 466616319891357 ns sh-tmu.0: ch0: used for clock events sh-tmu.0: ch0: used for periodic clock events sh-tmu.0: ch1: used as clock source clocksource: sh-tmu.0: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 183481014613 ns sched_clock: 32 bits at 250 Hz, resolution 4000000ns, wraps every 8589934590000000ns Calibrating delay loop (skipped)... 499.99 BogoMIPS PRESET (lpj=999999) pid_max: default: 32768 minimum: 301 Mount-cache hash table entries: 1024 (order: 0, 4096 bytes, linear) Mountpoint-cache hash table entries: 1024 (order: 0, 4096 bytes, linear) CPU: SH7724 Performance Events: sh4a support registered devtmpfs: initialized clocksource: jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 7645041785100000 ns futex hash table entries: 256 (order: -1, 3072 bytes, linear) pinctrl core: initialized pinctrl subsystem NET: Registered PF_NETLINK/PF_ROUTE protocol family DMA: preallocated 128 KiB GFP_KERNEL pool for atomic allocations vpu: forcing memory chunk size to 0x00800000 veu0: forcing memory chunk size to 0x00400000 sh-pfc pfc-sh7724: sh7724_pfc handling gpio 0 -> 179 sh-pfc pfc-sh7724: sh7724_pfc handling gpio 180 -> 486 sh-pfc pfc-sh7724: sh7724_pfc support registered HW Breakpoints: SH-4A UBC support registered sh-pfc pfc-sh7724: Use of GPIO API for function requests is deprecated, convert to pinctrl USB1 function is selected usbcore: registered new interface driver usbfs usbcore: registered new interface driver hub usbcore: registered new device driver usb mc: Linux media interface: v0.10 videodev: Linux video capture interface: v2.00 pps_core: LinuxPPS API ver. 1 registered pps_core: Software ver. 5.3.6 - Copyright 2005-2007 Rodolfo Giometti <giometti@linux.it> PTP clock support registered sh_cmt sh-cmt-32.0: kept as earlytimer sh_tmu sh-tmu.0: kept as earlytimer sh_tmu sh-tmu.1: ch0: used for clock events sh_tmu sh-tmu.1: ch1: used as clock source clocksource: sh-tmu.1: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 183481014613 ns Advanced Linux Sound Architecture Driver Initialized. clocksource: Switched to clocksource sh-tmu.0 sh_tmu sh-tmu.0: ch0: used for oneshot clock events NET: Registered PF_INET protocol family IP idents hash table entries: 4096 (order: 3, 32768 bytes, linear) tcp_listen_portaddr_hash hash table entries: 512 (order: 0, 4096 bytes, linear) TCP established hash table entries: 2048 (order: 1, 8192 bytes, linear) TCP bind hash table entries: 2048 (order: 1, 8192 bytes, linear) TCP: Hash tables configured (established 2048 bind 2048) UDP hash table entries: 256 (order: 0, 4096 bytes, linear) UDP-Lite hash table entries: 256 (order: 0, 4096 bytes, linear) sq: Registering store queue API. unable to read sh_eth MAC address workingset: timestamp_bits=14 max_order=16 bucket_order=2 squashfs: version 4.0 (2009/01/31) Phillip Lougher Console: switching to colour frame buffer device 100x30 sh_mobile_lcdc_fb sh_mobile_lcdc_fb.0: registered sh_mobile_lcdc_fb.0/mainlcd as 800x480 16bpp. SuperH (H)SCI(F) driver initialized sh-sci.0: ttySC0 at MMIO 0xffe00000 (irq = 80, base_baud = 0) is a scif printk: console [ttySC0] enabled sh-sci.1: ttySC1 at MMIO 0xffe10000 (irq = 81, base_baud = 0) is a scif sh-sci.2: ttySC2 at MMIO 0xffe20000 (irq = 82, base_baud = 0) is a scif sh-sci.3: ttySC3 at MMIO 0xa4e30000 (irq = 56, base_baud = 0) is a scifa sh-sci.4: ttySC4 at MMIO 0xa4e40000 (irq = 88, base_baud = 0) is a scifa sh-sci.5: ttySC5 at MMIO 0xa4e50000 (irq = 109, base_baud = 0) is a scifa physmap-flash physmap-flash.0: physmap platform flash device: [mem 0x00000000-0x03ffffff] physmap-flash.0: Found 1 x16 devices at 0x0 in 16-bit bank. Manufacturer ID 0x000001 Chip ID 0x002301 Amd/Fujitsu Extended Query Table at 0x0040 Amd/Fujitsu Extended Query version 1.3. number of CFI chips: 1 Creating 2 MTD partitions on "physmap-flash.0": 0x000000000000-0x000000500000 : "boot loader" 0x000000500000-0x000004000000 : "free-area" phram: rootfs device: 0x2000000 at 0x8a000000 for erasesize 0x1000 SMSC LAN8700: probe of sh7724-ether-0:1f failed with error -22 libphy: sh_mii: probed sh-eth sh7724-ether.0 eth0: Base address at 0xa4600000, 00:00:87:6c:1d:c2, IRQ 91. r8a66597_hcd r8a66597_hcd.0: USB Host Controller r8a66597_hcd r8a66597_hcd.0: new USB bus registered, assigned bus number 1 r8a66597_hcd r8a66597_hcd.0: irq 65, io base 0xa4d80000 usb usb1: New USB device found, idVendor=1d6b, idProduct=0002, bcdDevice= 5.15 usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1 usb usb1: Product: USB Host Controller usb usb1: Manufacturer: Linux 5.15.0-rc1 r8a66597_hcd usb usb1: SerialNumber: r8a66597_hcd hub 1-0:1.0: USB hub found hub 1-0:1.0: 1 port detected sh-rtc sh-rtc: registered as rtc0 sh-rtc sh-rtc: hctosys: unable to read the hardware clock renesas-ceu renesas-ceu.0: Renesas Capture Engine Unit renesas-ceu.0 renesas-ceu renesas-ceu.1: Renesas Capture Engine Unit renesas-ceu.1 cpufreq: SuperH CPU frequency driver. cpufreq: cpufreq_online: ->get() failed sh_mmcif sh_mmcif.0: Platform OCR mask is ignored heartbeat: version 0.1.2 loaded ALSA device list: No soundcards found. sh_mmcif sh_mmcif.0: Chip version 0x0000, clock rate 83MHz sh_mobile_sdhi sh_mobile_sdhi.0: mmc0 base at 0x04ce0000, max clock rate 83 MHz Waiting for root device /dev/mmcblk0... random: fast init done usb 1-1: new full-speed USB device number 2 using r8a66597_hcd mmc0: new SDHC card at address fb83 mmcblk0: mmc0:fb83 SE08G 7.22 GiB VFS: Mounted root (squashfs filesystem) readonly on device 179:0. devtmpfs: mounted Freeing unused kernel image (initmem) memory: 124K This architecture does not have kernel memory protection. Run /sbin/init as init process usb 1-1: New USB device found, idVendor=10c4, idProduct=ea60, bcdDevice= 1.00 usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 usb 1-1: Product: CP2102 USB to UART Bridge Controller usb 1-1: Manufacturer: Silicon Labs usb 1-1: SerialNumber: 0001 NET: Registered PF_UNIX/PF_LOCAL protocol family Starting syslogd: OK Starting klogd: OK Running sysctl: OK Starting mdev... OK usbcore: registered new interface driver cp210x usbserial: USB Serial support registered for cp210x cp210x 1-1:1.0: cp210x converter detected usb 1-1: cp210x converter now attached to ttyUSB0 Saving random seed: SKIP (read-only file system detected) Starting haveged: haveged: command socket is listening at fd 3 OK Starting network: Generic PHY sh7724-ether-0:1f: attached PHY driver (mii_bus:phy_addr=sh7724-ether-0:1f, irq=POLL) sh-eth sh7724-ether.0 eth0: Link is Up - 100Mbps/Full - flow control off udhcpc: started, v1.33.0 NET: Registered PF_PACKET protocol family udhcpc: sending discover udhcpc: sending select for 192.168.3.251 udhcpc: lease of 192.168.3.251 obtained, lease time 3600 deleting routers adding dns 192.168.3.1 adding dns 192.168.3.2 OK Starting openntpd: -s option no longer works and will be removed soon. Please reconfigure to use constraints or trusted servers. constraint configured without libtls support constraint configured without libtls support constraint configured without libtls support OK No persistent location to store SSH host keys. New keys will be generated at each boot. Are you sure this is what you want to do? Starting dropbear sshd: NET: Registered PF_INET6 protocol family Segment Routing with IPv6 In-situ OAM (IOAM) with IPv6 OK Starting nodectrl: OK Starting tlwbe_heartbeat: random: python3: uninitialized urandom read (24 bytes read) random: crng init done CN12 SD/MMC Vdd: incomplete constraints, leaving on OK Welcome to Buildroot ecovec24 login: sh4life $ cat /proc/cpuinfo machine : R0P7724 (EcoVec) processor : 0 cpu family : sh4a cpu type : SH7724 cut : 11.x cpu flags : fpu perfctr llsc l2 cache type : split (harvard) icache size : 32KiB (4-way) dcache size : 32KiB (4-way) scache size : 256KiB (4-way) address sizes : 29 bits physical bogomips : 499.99 $
On 9/15/21 8:50 AM, Geert Uytterhoeven wrote: > arch/sh/include/asm/pgtable-3level.h:37:9: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] ... > The truncation to "unsigned long" has been there since forever, so > probably it still works fine ;-) 1) Linux is LP64 so sizeof(long) and sizeof(pointer) always match, so it's not truncating. 2) The sh5 only ever shipped evaluation units, it never had a production run, and we haven't implemented j64 yet, so all superh targets are currently 32 bit. (I.E. it's complaining about _expanding_ the pointer, which shouldn't be a problem as long as endianness is respected.) Rob
Hi Rob, On Fri, Sep 17, 2021 at 3:12 AM Rob Landley <rob@landley.net> wrote: > On 9/15/21 8:50 AM, Geert Uytterhoeven wrote: > > arch/sh/include/asm/pgtable-3level.h:37:9: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] > ... > > The truncation to "unsigned long" has been there since forever, so > > probably it still works fine ;-) > > 1) Linux is LP64 so sizeof(long) and sizeof(pointer) always match, so it's not > truncating. > > 2) The sh5 only ever shipped evaluation units, it never had a production run, > and we haven't implemented j64 yet, so all superh targets are currently 32 bit. > (I.E. it's complaining about _expanding_ the pointer, which shouldn't be a > problem as long as endianness is respected.) The build error is for 32-bit. If CONFIG_X2TLB=y, pgd_t.pgd is "unsigned long long", i.e. 64-bit, so casting it to a pointer needs an intermediate cast to "unsigned long". See arch/sh/include/asm/page.h: #ifdef CONFIG_X2TLB typedef struct { unsigned long pte_low, pte_high; } pte_t; typedef struct { unsigned long long pgprot; } pgprot_t; typedef struct { unsigned long long pgd; } pgd_t; #define pte_val(x) \ ((x).pte_low | ((unsigned long long)(x).pte_high << 32)) #define __pte(x) \ ({ pte_t __pte = {(x), ((unsigned long long)(x)) >> 32}; __pte; }) #else typedef struct { unsigned long pte_low; } pte_t; typedef struct { unsigned long pgprot; } pgprot_t; typedef struct { unsigned long pgd; } pgd_t; #define pte_val(x) ((x).pte_low) #define __pte(x) ((pte_t) { (x) } ) #endif Gr{oetje,eeting}s, Geert
On 9/17/21 4:21 AM, Geert Uytterhoeven wrote: > Hi Rob, > > On Fri, Sep 17, 2021 at 3:12 AM Rob Landley <rob@landley.net> wrote: >> On 9/15/21 8:50 AM, Geert Uytterhoeven wrote: >> > arch/sh/include/asm/pgtable-3level.h:37:9: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] >> ... >> > The truncation to "unsigned long" has been there since forever, so >> > probably it still works fine ;-) >> >> 1) Linux is LP64 so sizeof(long) and sizeof(pointer) always match, so it's not >> truncating. >> >> 2) The sh5 only ever shipped evaluation units, it never had a production run, >> and we haven't implemented j64 yet, so all superh targets are currently 32 bit. >> (I.E. it's complaining about _expanding_ the pointer, which shouldn't be a >> problem as long as endianness is respected.) > > The build error is for 32-bit. I know. I'm just saying your fix doesn't seem wrong. (Personally I would have just cast once to (void *) and let C not being C++ do the right thing, but eh...) Not sure I understand the point of the original code though. Under what conditions do you want 64 bit page tables on a 32 bit system? (Some PAE variant?) > If CONFIG_X2TLB=y, pgd_t.pgd is "unsigned long long", i.e. 64-bit, so > casting it to a pointer needs an intermediate cast to "unsigned long". > See arch/sh/include/asm/page.h: I can't find a user of it. None of the existing defconfigs select this option: $ grep -r X2TLB arch/sh/configs And it's weird in other ways, like 8k page size. (Why...?) Google finds sh-x2 mentioned in Renesas' sh4a debugger docs, but I've never played with the renesas "a" variants. The stuff I'm familiar with is all following up on what Hitachi did, not Renesas. *shrug* You fix looks good enough to me. Acked-by: Rob Landley <rob@landley.net> Rob
Hi Rob, On Sat, Sep 18, 2021 at 6:45 AM Rob Landley <rob@landley.net> wrote: > On 9/17/21 4:21 AM, Geert Uytterhoeven wrote: > > On Fri, Sep 17, 2021 at 3:12 AM Rob Landley <rob@landley.net> wrote: > >> On 9/15/21 8:50 AM, Geert Uytterhoeven wrote: > >> > arch/sh/include/asm/pgtable-3level.h:37:9: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] > >> ... > >> > The truncation to "unsigned long" has been there since forever, so > >> > probably it still works fine ;-) > >> > >> 1) Linux is LP64 so sizeof(long) and sizeof(pointer) always match, so it's not > >> truncating. > >> > >> 2) The sh5 only ever shipped evaluation units, it never had a production run, > >> and we haven't implemented j64 yet, so all superh targets are currently 32 bit. > >> (I.E. it's complaining about _expanding_ the pointer, which shouldn't be a > >> problem as long as endianness is respected.) > > > > The build error is for 32-bit. > > I know. I'm just saying your fix doesn't seem wrong. (Personally I would have > just cast once to (void *) and let C not being C++ do the right thing, but eh...) That would still give a warning, as "unsigned long long" us 64-bit, and "void *" is 32-bit. > Not sure I understand the point of the original code though. Under what > conditions do you want 64 bit page tables on a 32 bit system? (Some PAE variant?) > > > If CONFIG_X2TLB=y, pgd_t.pgd is "unsigned long long", i.e. 64-bit, so > > casting it to a pointer needs an intermediate cast to "unsigned long". > > See arch/sh/include/asm/page.h: > > I can't find a user of it. None of the existing defconfigs select this option: > > $ grep -r X2TLB arch/sh/configs Note that these are reduced config files, so auto-selected symbols are not present in the defconfig. config X2TLB def_bool y depends on (CPU_SHX2 || CPU_SHX3) && MMU and e.g. config CPU_SUBTYPE_SH7723 bool "Support SH7723 processor" select CPU_SH4A select CPU_SHX2 > > And it's weird in other ways, like 8k page size. (Why...?) > > Google finds sh-x2 mentioned in Renesas' sh4a debugger docs, but I've never > played with the renesas "a" variants. The stuff I'm familiar with is all > following up on what Hitachi did, not Renesas. > > *shrug* You fix looks good enough to me. > > Acked-by: Rob Landley <rob@landley.net> Thanks! Gr{oetje,eeting}s, Geert
Hi Geert! On 9/15/21 15:50, Geert Uytterhoeven wrote: > If X2TLB=y (CPU_SHX2=y or CPU_SHX3=y, e.g. migor_defconfig), pgd_t.pgd > is "unsigned long long", causing: > > In file included from arch/sh/include/asm/pgtable.h:13, > from include/linux/pgtable.h:6, > from include/linux/mm.h:33, > from arch/sh/kernel/asm-offsets.c:14: > arch/sh/include/asm/pgtable-3level.h: In function ‘pud_pgtable’: > arch/sh/include/asm/pgtable-3level.h:37:9: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] > 37 | return (pmd_t *)pud_val(pud); > | ^ > > Fix this by adding an intermediate cast to "unsigned long", which is > basically what the old code did before. Thanks a lot for fixing this. My SH kernel build becomes much less noisy with your patch. I can also confirm that it doesn't break the boot on my SH-7785LCR as expected. Tested-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> Adrian
diff --git a/arch/sh/include/asm/pgtable-3level.h b/arch/sh/include/asm/pgtable-3level.h index 56bf35c2f29c2b83..cdced80a7ffa3944 100644 --- a/arch/sh/include/asm/pgtable-3level.h +++ b/arch/sh/include/asm/pgtable-3level.h @@ -34,7 +34,7 @@ typedef struct { unsigned long long pmd; } pmd_t; static inline pmd_t *pud_pgtable(pud_t pud) { - return (pmd_t *)pud_val(pud); + return (pmd_t *)(unsigned long)pud_val(pud); } /* only used by the stubbed out hugetlb gup code, should never be called */
If X2TLB=y (CPU_SHX2=y or CPU_SHX3=y, e.g. migor_defconfig), pgd_t.pgd is "unsigned long long", causing: In file included from arch/sh/include/asm/pgtable.h:13, from include/linux/pgtable.h:6, from include/linux/mm.h:33, from arch/sh/kernel/asm-offsets.c:14: arch/sh/include/asm/pgtable-3level.h: In function ‘pud_pgtable’: arch/sh/include/asm/pgtable-3level.h:37:9: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] 37 | return (pmd_t *)pud_val(pud); | ^ Fix this by adding an intermediate cast to "unsigned long", which is basically what the old code did before. Fixes: 9cf6fa2458443118 ("mm: rename pud_page_vaddr to pud_pgtable and make it return pmd_t *") Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> --- Compile-tested only, as the remote Migo-R is MIA. Anyone with e.g. a Migo-R or Ecovec to boot-test? The truncation to "unsigned long" has been there since forever, so probably it still works fine ;-) ... once the other build issues are fixed. --- arch/sh/include/asm/pgtable-3level.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)