* Build regressions/improvements in v3.14-rc3
@ 2014-02-18 9:02 Geert Uytterhoeven
2014-02-18 9:06 ` Geert Uytterhoeven
` (2 more replies)
0 siblings, 3 replies; 31+ messages in thread
From: Geert Uytterhoeven @ 2014-02-18 9:02 UTC (permalink / raw)
To: linux-kernel
Below is the list of build error/warning regressions/improvements in
v3.14-rc3[1] compared to v3.13[2].
Summarized:
- build errors: +21/-31
- build warnings: +188/-74
JFYI, when comparing v3.14-rc3[1] to v3.14-rc2[3], the summaries are:
- build errors: +17/-6
- build warnings: +114/-57
As I haven't mastered kup yet, there's no verbose summary at
http://www.kernel.org/pub/linux/kernel/people/geert/linux-log/v3.14-rc3.summary.gz
Happy fixing! ;-)
Thanks to the linux-next team for providing the build service.
[1] http://kisskb.ellerman.id.au/kisskb/head/7190/ (all 119 configs)
[2] http://kisskb.ellerman.id.au/kisskb/head/7082/ (all 119 configs)
[3] http://kisskb.ellerman.id.au/kisskb/head/7167/ (all 119 configs)
*** ERRORS ***
21 regressions:
+ /scratch/kisskb/src/arch/powerpc/include/asm/floppy.h: error: 'isa_bridge_pcidev' undeclared (first use in this function): => 142:20
+ /scratch/kisskb/src/arch/powerpc/kernel/exceptions-64s.S: Error: attempt to move .org backwards: => 1312
+ /scratch/kisskb/src/drivers/gpio/gpio-xtensa.c: Error: unknown opcode or format name 'read_impwire': => 87
+ /scratch/kisskb/src/drivers/gpio/gpio-xtensa.c: Error: unknown opcode or format name 'rur.expstate': => 110
+ /scratch/kisskb/src/drivers/gpio/gpio-xtensa.c: Error: unknown opcode or format name 'wrmsk_expstate': => 124
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: #error m8xx_pcmcia: Bad configuration!: => 127:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: 'PCMCIA_IO_WIN_NO' undeclared here (not in a function): => 220:30
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: 'PCMCIA_MEM_WIN_NO' undeclared here (not in a function): => 219:32
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: 'PCMCIA_SOCKETS_NO' undeclared here (not in a function): => 225:34
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: 'irq_map' undeclared (first use in this function): => 1027:10
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: 's' undeclared (first use in this function): => 982:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: expected ')' before 'PCMCIA_BOARD_MSG': => 1036:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit declaration of function 'hardware_disable' [-Werror=implicit-function-declaration]: => 1137:3
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit declaration of function 'hardware_enable' [-Werror=implicit-function-declaration]: => 1080:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit declaration of function 'mk_int_int_mask' [-Werror=implicit-function-declaration]: => 748:6
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit declaration of function 'socket_get' [-Werror=implicit-function-declaration]: => 657:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit declaration of function 'voltage_set' [-Werror=implicit-function-declaration]: => 686:2
+ /scratch/kisskb/src/drivers/usb/gadget/f_fs.c: error: 'ffsusb_func' undeclared (first use in this function): => 2688:1
+ error: jump_label.c: undefined reference to `arch_jump_label_transform': => .text+0x85e5c)
+ error: kvm_main.c: undefined reference to `__stack_chk_guard': => .text+0x6895)
+ error: vsprintf.c: relocation truncated to fit: R_AVR32_9UW_PCREL against `.text'+bc8: => (.text+0x9c2)
31 improvements:
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'dmac_resource.end'): 87:2 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'dmac_resource.start'): 86:2 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'epbr_resources[0].end'): 69:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'epbr_resources[0].start'): 68:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'epbr_resources[1].end'): 74:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'epbr_resources[1].start'): 73:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'femi_resources[0].end'): 50:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'femi_resources[0].start'): 49:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'femi_resources[1].end'): 55:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'femi_resources[1].start'): 54:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'pbr_resources[0].end'): 100:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'pbr_resources[0].start'): 99:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'pbr_resources[1].end'): 105:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: (near initialization for 'pbr_resources[1].start'): 104:3 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: braced-group within expression allowed only inside a function: 30:12, 73:12, 31:10, 99:12, 68:12, 36:10, 86:11, 54:12, 74:10, 55:10, 104:12, 49:12, 69:10, 87:9, 105:10, 35:12, 50:10, 100:10 =>
- /scratch/kisskb/src/arch/sh/drivers/superhyway/ops-sh4-202.c: error: initializer element is not constant: 74:3, 86:2, 55:3, 105:3, 73:3, 68:3, 100:3, 104:3, 99:3, 87:2, 50:3, 69:3, 49:3, 54:3 =>
- /scratch/kisskb/src/arch/sh/kernel/kgdb.c: error: 'SIGTRAP' undeclared (first use in this function): 310:27 =>
- /scratch/kisskb/src/arch/sh/kernel/kgdb.c: error: dereferencing pointer to incomplete type: 243:22, 242:23 =>
- /scratch/kisskb/src/arch/sh/kernel/kgdb.c: error: implicit declaration of function 'task_stack_page' [-Werror=implicit-function-declaration]: 225:32 =>
- /scratch/kisskb/src/arch/sh/mm/cache-sh4.c: error: 'cached_to_uncached' undeclared (first use in this function): 99:17 =>
- /scratch/kisskb/src/arch/sh/mm/cache-sh4.c: error: implicit declaration of function 'cpu_context' [-Werror=implicit-function-declaration]: 192:2 =>
- /scratch/kisskb/src/drivers/staging/comedi/drivers/das6402.c: error: implicit declaration of function 'outw_p' [-Werror=implicit-function-declaration]: 164:3 =>
- /scratch/kisskb/src/drivers/staging/speakup/speakup_dtlk.c: error: implicit declaration of function 'inw_p' [-Werror=implicit-function-declaration]: 362:2 =>
- /scratch/kisskb/src/drivers/tty/ehv_bytechan.c: error: type defaults to 'int' in declaration of 'console_initcall' [-Werror=implicit-int]: 363:1 =>
- /scratch/kisskb/src/kernel/bounds.c: error: -mcall-aixdesc must be big endian: 1:0 =>
- /scratch/kisskb/src/scripts/mod/devicetable-offsets.c: error: -mcall-aixdesc must be big endian: 1:0 =>
- /scratch/kisskb/src/scripts/mod/empty.c: error: -mcall-aixdesc must be big endian: 1:0 =>
- <stdin>: error: -mcall-aixdesc must be big endian: 1:0 =>
- error: nsc-ircc.c: relocation truncated to fit: R_PPC64_REL24 against symbol `.printk' defined in .text.unlikely section in kernel/built-in.o: (.text+0x1ffa5f4), (.text+0x1ffa634), (.text+0x1ff8e2c), (.text+0x1ffa674), (.text+0x1ffa5b0), (.text+0x1ff8c90), (.text+0x1ff99d8), (.text+0x1ffa6b4), (.text+0x1ffa6f8) =>
- error: nsc-ircc.c: relocation truncated to fit: R_PPC64_REL24 against symbol `_savegpr0_25' defined in .text.save.restore section in arch/powerpc/lib/built-in.o: (.text+0x1ff997c) =>
- error: vsprintf.c: relocation truncated to fit: R_AVR32_9UW_PCREL against `.text'+b8c: (.text+0x986) =>
*** WARNINGS ***
188 regressions:
+ /scratch/kisskb/src/arch/arm/kernel/return_address.c: warning: #warning "TODO: return_address should use unwind tables" [-Wcpp]: => 63:2
+ /scratch/kisskb/src/arch/cris/include/arch-v10/arch/system.h: warning: "nop" redefined [enabled by default]: => 39:0
+ /scratch/kisskb/src/arch/x86/include/asm/io.h: warning: 'r.x[0]' may be used uninitialized in this function [-Wuninitialized]: => 67:1
+ /scratch/kisskb/src/arch/x86/kvm/pmu.c: warning: 'config' may be used uninitialized in this function [-Wuninitialized]: => 245:19
+ /scratch/kisskb/src/arch/x86/kvm/x86.c: warning: 'exchanged' may be used uninitialized in this function [-Wuninitialized]: => 4394:2
+ /scratch/kisskb/src/drivers/base/memory.c: warning: 'section_nr' may be used uninitialized in this function [-Wuninitialized]: => 607:25
+ /scratch/kisskb/src/drivers/base/regmap/regmap.c: warning: 'ret' may be used uninitialized in this function [-Wuninitialized]: => 2181:9, 1799:6
+ /scratch/kisskb/src/drivers/block/nvme-core.c: warning: 'bvprv.bv_len' may be used uninitialized in this function [-Wuninitialized]: => 514:18
+ /scratch/kisskb/src/drivers/block/nvme-core.c: warning: 'bvprv.bv_offset' may be used uninitialized in this function [-Wuninitialized]: => 514:18
+ /scratch/kisskb/src/drivers/block/nvme-core.c: warning: 'bvprv.bv_page' may be used uninitialized in this function [-Wuninitialized]: => 511:17
+ /scratch/kisskb/src/drivers/gpio/gpio-sch311x.c: warning: ignoring return value of 'gpiochip_remove', declared with attribute warn_unused_result [-Wunused-result]: => 286:18
+ /scratch/kisskb/src/drivers/gpio/gpiolib.c: warning: unused variable 'gpio' [-Wunused-variable]: => 204:12
+ /scratch/kisskb/src/drivers/gpio/gpiolib.c: warning: unused variable 'gpio': => 204
+ /scratch/kisskb/src/drivers/gpu/drm/drm_crtc.c: warning: control reaches end of non-void function [-Wreturn-type]: => 697:1
+ /scratch/kisskb/src/drivers/gpu/drm/radeon/radeon_ttm.c: warning: comparison of distinct pointer types lacks a cast [enabled by default]: => 938:22
+ /scratch/kisskb/src/drivers/md/bcache/bset.c: warning: format '%li' expects argument of type 'long int', but argument 3 has type 'int' [-Wformat]: => 27:10
+ /scratch/kisskb/src/drivers/media/media-entity.c: warning: 'media_entity_pipeline_start' uses dynamic stack allocation [enabled by default]: => 322:1
+ /scratch/kisskb/src/drivers/mfd/max14577.c: warning: 'max14577_resume' defined but not used [-Wunused-function]: => 199:12
+ /scratch/kisskb/src/drivers/mfd/max14577.c: warning: 'max14577_suspend' defined but not used [-Wunused-function]: => 176:12
+ /scratch/kisskb/src/drivers/mfd/sec-core.c: warning: 'sec_pmic_resume' defined but not used [-Wunused-function]: => 340:12
+ /scratch/kisskb/src/drivers/mfd/sec-core.c: warning: 'sec_pmic_suspend' defined but not used [-Wunused-function]: => 318:12
+ /scratch/kisskb/src/drivers/misc/bh1770glc.c: warning: 'mode' may be used uninitialized in this function [-Wuninitialized]: => 236:29
+ /scratch/kisskb/src/drivers/misc/ics932s401.c: warning: 'x' may be used uninitialized in this function [-Wuninitialized]: 384:2 => 384:2, 384:16
+ /scratch/kisskb/src/drivers/mtd/nand/sh_flctl.c: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]: => 164:17, 181:17
+ /scratch/kisskb/src/drivers/net/ethernet/cadence/macb.c: warning: 'macb_resume' defined but not used [-Wunused-function]: => 2030:12
+ /scratch/kisskb/src/drivers/net/ethernet/cadence/macb.c: warning: 'macb_suspend' defined but not used [-Wunused-function]: => 2013:12
+ /scratch/kisskb/src/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]: => 303:11, 307:11
+ /scratch/kisskb/src/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]: => 467:46
+ /scratch/kisskb/src/drivers/net/ethernet/sfc/ef10.c: warning: 'efx_ef10_free_vis' uses dynamic stack allocation [enabled by default]: => 295:1
+ /scratch/kisskb/src/drivers/net/ethernet/sfc/mcdi.c: warning: '_efx_mcdi_rpc_finish' uses dynamic stack allocation [enabled by default]: => 617:1
+ /scratch/kisskb/src/drivers/net/ethernet/sfc/mcdi.c: warning: 'efx_mcdi_complete_async' uses dynamic stack allocation [enabled by default]: => 468:1
+ /scratch/kisskb/src/drivers/net/ethernet/sfc/mcdi.c: warning: 'efx_mcdi_read_assertion' uses dynamic stack allocation [enabled by default]: => 1425:1
+ /scratch/kisskb/src/drivers/net/ethernet/sfc/ptp.c: warning: 'efx_ptp_disable' uses dynamic stack allocation [enabled by default]: => 614:1
+ /scratch/kisskb/src/drivers/net/ethernet/sfc/ptp.c: warning: 'efx_ptp_enable' uses dynamic stack allocation [enabled by default]: => 591:1
+ /scratch/kisskb/src/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c: warning: 'stmmac_pltfr_resume' defined but not used [-Wunused-function]: => 295:12
+ /scratch/kisskb/src/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c: warning: 'stmmac_pltfr_suspend' defined but not used [-Wunused-function]: => 281:12
+ /scratch/kisskb/src/drivers/net/tun.c: warning: 'copylen' may be used uninitialized in this function [-Wuninitialized]: => 1085:21
+ /scratch/kisskb/src/drivers/net/tun.c: warning: 'linear' may be used uninitialized in this function [-Wuninitialized]: => 1085:6
+ /scratch/kisskb/src/drivers/net/usb/sr9800.c: warning: format '%ld' expects argument of type 'long int', but argument 5 has type 'size_t' [-Wformat]: => 826:2
+ /scratch/kisskb/src/drivers/net/virtio_net.c: warning: 'v' may be used uninitialized in this function [-Wuninitialized]: => 1372:4
+ /scratch/kisskb/src/drivers/net/wireless/wl3501_cs.c: warning: 'addr' may be used uninitialized in this function [-Wuninitialized]: 805:23 => 334:24, 804:23
+ /scratch/kisskb/src/drivers/net/wireless/wl3501_cs.c: warning: 'implemented' may be used uninitialized in this function [-Wuninitialized]/opt/cross/gcc-4.6-nolibc/sh4-linux/bin/sh4-linux-ld: : => 1763:5
+ /scratch/kisskb/src/drivers/net/wireless/wl3501_cs.c: warning: 'next' may be used uninitialized in this function [-Wuninitialized]: 343:24 => 343:24, 338:8
+ /scratch/kisskb/src/drivers/net/wireless/wl3501_cs.c: warning: 'restricted' may be used uninitialized in this function [-Wuninitialized]/opt/cross/gcc-4.6-nolibc/sh4-linux/bin/sh4-linux-ld: failed to merge target specific data of file net/netrom/nr_loopback.o: => 1771:23
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: warning: 'pending_events' defined but not used [-Wunused-variable]: => 408:12
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: warning: 'socket' defined but not used [-Wunused-variable]: => 225:27
+ /scratch/kisskb/src/drivers/platform/goldfish/goldfish_pipe.c: warning: 'status' may be used uninitialized in this function [-Wuninitialized]: => 334:3
+ /scratch/kisskb/src/drivers/power/bq24190_charger.c: warning: 'en_term' may be used uninitialized in this function [-Wuninitialized]: => 624:27
+ /scratch/kisskb/src/drivers/power/bq24190_charger.c: warning: 'force_20pct' may be used uninitialized in this function [-Wuninitialized]: => 617:27
+ /scratch/kisskb/src/drivers/power/bq27x00_battery.c: warning: 'bq27x00_powersupply_init' defined but not used [-Wunused-function]: => 693:12
+ /scratch/kisskb/src/drivers/power/bq27x00_battery.c: warning: 'bq27x00_powersupply_unregister' defined but not used [-Wunused-function]: => 724:13
+ /scratch/kisskb/src/drivers/power/da9052-battery.c: warning: 'level_lower' may be used uninitialized in this function [-Wuninitialized]: => 325:6
+ /scratch/kisskb/src/drivers/power/da9052-battery.c: warning: 'level_upper' may be used uninitialized in this function [-Wuninitialized]: => 324:22
+ /scratch/kisskb/src/drivers/power/da9052-battery.c: warning: 'vbat_lower' may be used uninitialized in this function [-Wuninitialized]: => 325:37
+ /scratch/kisskb/src/drivers/power/da9052-battery.c: warning: 'vbat_upper' may be used uninitialized in this function [-Wuninitialized]: => 324:59
+ /scratch/kisskb/src/drivers/power/ds2782_battery.c: warning: 'raw' may be used uninitialized in this function [-Wuninitialized]: => 213:17, 184:21, 200:21, 160:21, 144:20, 172:12, 115:16
+ /scratch/kisskb/src/drivers/power/ds2782_battery.c: warning: 'sense_res_raw' may be used uninitialized in this function [-Wuninitialized]: => 137:19
+ /scratch/kisskb/src/drivers/regulator/tps80031-regulator.c: warning: 'smps_mult' may be used uninitialized in this function [-Wuninitialized]: => 653:33
+ /scratch/kisskb/src/drivers/regulator/tps80031-regulator.c: warning: 'smps_offset' may be used uninitialized in this function [-Wuninitialized]: => 652:34
+ /scratch/kisskb/src/drivers/regulator/wm8350-regulator.c: warning: 'setting' may be used uninitialized in this function [-Wuninitialized]: => 133:19
+ /scratch/kisskb/src/drivers/rtc/rtc-cmos.c: warning: 'cmos_resume' defined but not used [-Wunused-function]: => 891:12
+ /scratch/kisskb/src/drivers/scsi/hpsa.c: warning: 'driver_support' is used uninitialized in this function [-Wuninitialized]: => 4375:17
+ /scratch/kisskb/src/drivers/spi/spi-atmel.c: warning: format '%x' expects argument of type 'unsigned int', but argument 7 has type 'dma_addr_t' [-Wformat]: => 1246:3
+ /scratch/kisskb/src/drivers/spi/spi-atmel.c: warning: format '%x' expects argument of type 'unsigned int', but argument 9 has type 'dma_addr_t' [-Wformat]: => 1246:3
+ /scratch/kisskb/src/drivers/staging/dgap/dgap_driver.c: warning: the frame size of 1032 bytes is larger than 1024 bytes [-Wframe-larger-than=]: => 961:1
+ /scratch/kisskb/src/drivers/staging/dgap/dgap_driver.c: warning: the frame size of 1064 bytes is larger than 1024 bytes [-Wframe-larger-than=]: => 961:1
+ /scratch/kisskb/src/drivers/staging/dgap/dgap_tty.c: warning: the frame size of 1064 bytes is larger than 1024 bytes [-Wframe-larger-than=]: => 658:1
+ /scratch/kisskb/src/drivers/staging/goldfish/goldfish_nand.c: warning: 'rv' may be used uninitialized in this function [-Wuninitialized]: => 94:2
+ /scratch/kisskb/src/drivers/staging/iio/adc/mxs-lradc.c: warning: comparison of distinct pointer types lacks a cast [enabled by default]: => 1614:6
+ /scratch/kisskb/src/drivers/staging/iio/adc/mxs-lradc.c: warning: passing argument 1 of '__div64_32' from incompatible pointer type [enabled by default]: => 1614:6
+ /scratch/kisskb/src/drivers/staging/iio/adc/mxs-lradc.c: warning: right shift count >= width of type [enabled by default]: => 1614:6
+ /scratch/kisskb/src/drivers/target/target_core_alua.c: warning: 'alua_ascq' may be used uninitialized in this function [-Wuninitialized]: => 773:18
+ /scratch/kisskb/src/drivers/uio/uio.c: warning: 'ret' may be used uninitialized in this function [-Wuninitialized]: => 269:6
+ /scratch/kisskb/src/drivers/usb/gadget/f_fs.c: warning: 'data_len' may be used uninitialized in this function [-Wuninitialized]: => 665:17
+ /scratch/kisskb/src/drivers/usb/gadget/f_fs.c: warning: 'kernel_ffsusb_func' defined but not used [-Wunused-variable]: => 2688:1
+ /scratch/kisskb/src/drivers/usb/gadget/f_fs.c: warning: control reaches end of non-void function [-Wreturn-type]: => 2688:1
+ /scratch/kisskb/src/drivers/video/backlight/adp8860_bl.c: warning: 'reg_val' may be used uninitialized in this function [-Wuninitialized]: => 624:11, 568:10
+ /scratch/kisskb/src/drivers/video/backlight/adp8870_bl.c: warning: 'reg_val' may be used uninitialized in this function [-Wuninitialized]: => 809:11
+ /scratch/kisskb/src/drivers/video/ocfb.c: warning: passing argument 3 of 'dma_alloc_attrs' from incompatible pointer type [enabled by default]: => 335:2
+ /scratch/kisskb/src/drivers/w1/slaves/w1_therm.c: warning: 'verdict' may be used uninitialized in this function [-Wuninitialized]: => 255:2
+ /scratch/kisskb/src/fs/autofs4/inode.c: warning: 'pgrp' may be used uninitialized in this function: => 213
+ /scratch/kisskb/src/fs/btrfs/hash.c: warning: 'btrfs_crc32c' uses dynamic stack allocation [enabled by default]: => 50:1
+ /scratch/kisskb/src/fs/ceph/acl.c: warning: 'name' may be used uninitialized in this function [-Wuninitialized]: => 80:7
+ /scratch/kisskb/src/fs/f2fs/segment.c: warning: '__next_free_blkoff' uses dynamic stack allocation [enabled by default]: => 683:1
+ /scratch/kisskb/src/fs/f2fs/segment.c: warning: 'add_discard_addrs' uses dynamic stack allocation [enabled by default]: => 318:1
+ /scratch/kisskb/src/fs/lockd/svc.c: warning: label 'err_sysctl' defined but not used [-Wunused-label]: => 625:1
+ /scratch/kisskb/src/fs/ncpfs/dir.c: warning: unused variable 'sb' [-Wunused-variable]: => 134:23
+ /scratch/kisskb/src/fs/notify/fanotify/fanotify.c: warning: 'marks_ignored_mask' may be used uninitialized in this function [-Wuninitialized]: 214:32 => 139:2, 139:32
+ /scratch/kisskb/src/fs/notify/fanotify/fanotify.c: warning: 'marks_mask' may be used uninitialized in this function [-Wuninitialized]: 214:17 => 139:2, 139:17
+ /scratch/kisskb/src/fs/posix_acl.c: warning: control reaches end of non-void function [-Wreturn-type]: => 34:1
+ /scratch/kisskb/src/fs/xfs/./xfs_trace.h: warning: format '%llu' expects argument of type 'long long unsigned int', but argument 6 has type 'xfs_fsblock_t' [-Wformat]: => 674:1
+ /scratch/kisskb/src/include/linux/file.h: warning: 'output.file' may be used uninitialized in this function [-Wuninitialized]: => 37:7
+ /scratch/kisskb/src/include/linux/file.h: warning: 'output.need_put' may be used uninitialized in this function [-Wuninitialized]: => 36:2
+ /scratch/kisskb/src/include/linux/list.h: warning: 'head' may be used uninitialized in this function [-Wuninitialized]: => 643:11
+ /scratch/kisskb/src/include/linux/list_nulls.h: warning: 'node' may be used uninitialized in this function [-Wuninitialized]: => 35:2
+ /scratch/kisskb/src/include/linux/mtd/map.h: warning: 'r.x[0]' may be used uninitialized in this function [-Wuninitialized]: 279:4, 349:2, 324:17, 278:6, 367:14, 290:22, 414:2, 332:11, 397:11 => 279:4, 349:2, 324:17, 278:6, 367:14, 290:22, 414:2, 278:3, 332:11, 397:11
+ /scratch/kisskb/src/include/linux/spinlock.h: warning: 'dst_ptl' may be used uninitialized in this function [-Wuninitialized]: => 343:2
+ /scratch/kisskb/src/include/linux/spinlock.h: warning: 'ptl' may be used uninitialized in this function [-Wuninitialized]: => 343:2
+ /scratch/kisskb/src/include/linux/spinlock.h: warning: 'vf' may be used uninitialized in this function [-Wuninitialized]: => 343:2
+ /scratch/kisskb/src/include/linux/sunrpc/cache.h: warning: 'rv' may be used uninitialized in this function [-Wuninitialized]: => 295:2
+ /scratch/kisskb/src/include/uapi/linux/swab.h: warning: 'r.x[0]' may be used uninitialized in this function [-Wuninitialized]: => 60:2
+ /scratch/kisskb/src/kernel/seccomp.c: warning: 'ret' may be used uninitialized in this function [-Wuninitialized]: => 455:15
+ /scratch/kisskb/src/kernel/seccomp.c: warning: control reaches end of non-void function [-Wreturn-type]: => 112:1
+ /scratch/kisskb/src/kernel/trace/ftrace.c: warning: 'control_ops_free' defined but not used [-Wunused-function]: => 240:13
+ /scratch/kisskb/src/kernel/trace/trace_events.c: warning: 'flags' may be used uninitialized in this function [-Wuninitialized]: => 709:2
+ /scratch/kisskb/src/lib/debugobjects.c: warning: 'obj_states' defined but not used [-Wunused-variable]: => 73:20
+ /scratch/kisskb/src/lib/debugobjects.c: warning: unused variable 'hint' [-Wunused-variable]: => 254:9
+ /scratch/kisskb/src/lib/lz4/lz4hc_compress.c: warning: 'delta' may be used uninitialized in this function [-Wuninitialized]: => 174:20
+ /scratch/kisskb/src/lib/mpi/mpi-inline.h: warning: '______f' is static but declared in inline function 'mpihelp_add' which is not static [enabled by default]: => 70:2, 73:2
+ /scratch/kisskb/src/lib/mpi/mpi-inline.h: warning: '______f' is static but declared in inline function 'mpihelp_add_1' which is not static [enabled by default]: => 45:2, 49:4, 56:2
+ /scratch/kisskb/src/lib/mpi/mpi-inline.h: warning: '______f' is static but declared in inline function 'mpihelp_sub' which is not static [enabled by default]: => 113:2, 116:2
+ /scratch/kisskb/src/lib/mpi/mpi-inline.h: warning: '______f' is static but declared in inline function 'mpihelp_sub_1' which is not static [enabled by default]: => 92:4, 99:2, 88:2
+ /scratch/kisskb/src/lib/vsprintf.c: warning: 'resource_string.isra.11' uses dynamic stack allocation [enabled by default]: => 744:1
+ /scratch/kisskb/src/mm/memblock.c: warning: 'r' may be used uninitialized in this function [-Wuninitialized]: => 944:15
+ /scratch/kisskb/src/mm/memcontrol.c: warning: 'thresholds' may be used uninitialized in this function [-Wuninitialized]: 5683:19, 5801:32, 5783:35 => 5823:35, 5713:19, 5710:16, 5841:32
+ /scratch/kisskb/src/mm/memory.c: warning: #warning Unfortunate NUMA and NUMA Balancing config, growing page-frame for last_cpupid. [-Wcpp]: => 74:2
+ /scratch/kisskb/src/mm/percpu.c: warning: format '%lx' expects argument of type 'long unsigned int', but argument 3 has type 'unsigned int' [-Wformat]: => 1693:3
+ /scratch/kisskb/src/net/ceph/osdmap.c: warning: 'crush_do_rule_ary' uses dynamic stack allocation [enabled by default]: => 1153:1
+ /scratch/kisskb/src/net/core/neighbour.c: warning: 'p_name' may be used uninitialized in this function [-Wuninitialized]: => 3094:10
+ /scratch/kisskb/src/net/core/skbuff.c: warning: 'err' may be used uninitialized in this function [-Wuninitialized]: => 1340:6
+ /scratch/kisskb/src/net/ipv4/inet_connection_sock.c: warning: 'head' may be used uninitialized in this function [-Wuninitialized]: => 209:2
+ /scratch/kisskb/src/net/ipv4/ip_options.c: warning: 'faddr' may be used uninitialized in this function [-Wuninitialized]: => 183:16
+ /scratch/kisskb/src/net/ipv4/ip_options.c: warning: 'nexthop' may be used uninitialized in this function [-Wuninitialized]: => 656:16
+ /scratch/kisskb/src/net/ipv4/ping.c: warning: 'code' may be used uninitialized in this function [-Wuninitialized]: 450:63, 533:37 => 457:63, 552:42, 540:37
+ /scratch/kisskb/src/net/ipv4/ping.c: warning: 'icmph' may be used uninitialized in this function [-Wuninitialized]: 487:48, 487:35, 494:29 => 494:2, 501:29, 494:48, 494:35
+ /scratch/kisskb/src/net/ipv4/ping.c: warning: 'type' may be used uninitialized in this function [-Wuninitialized]: 490:2, 664:21, 451:29, 506:3, 545:42 => 497:2, 671:21, 458:29, 513:3, 552:42, 671:2
+ /scratch/kisskb/src/net/ipv4/tcp.c: warning: 'copy' may be used uninitialized in this function [-Wuninitialized]: => 927:3
+ /scratch/kisskb/src/net/ipv4/udp.c: warning: 'sk' may be used uninitialized in this function [-Wuninitialized]: => 1904:8
+ /scratch/kisskb/src/net/nfc/llcp_core.c: warning: 'sn_len' may be used uninitialized in this function [-Wuninitialized]: => 801:12
+ /scratch/kisskb/src/net/openvswitch/datapath.c: warning: 'err' may be used uninitialized in this function [-Wuninitialized]: => 401:6
+ /scratch/kisskb/src/net/rds/rdma.c: warning: 'iov_size' may be used uninitialized in this function [-Wuninitialized]: => 697:15
+ /scratch/kisskb/src/net/sched/sch_sfb.c: warning: 'salt' may be used uninitialized in this function [-Wuninitialized]: => 313:12
+ /scratch/kisskb/src/net/sunrpc/svcauth_unix.c: warning: 'gid' may be used uninitialized in this function [-Wuninitialized]: => 517:19
+ /scratch/kisskb/src/net/sunrpc/svcauth_unix.c: warning: 'id' may be used uninitialized in this function [-Wuninitialized]: => 495:17
+ /scratch/kisskb/src/net/sunrpc/svcsock.c: warning: 'optname' may be used uninitialized in this function [-Wuninitialized] CC drivers/net/ppp/ppp_deflate.o: => 725:6
+ /scratch/kisskb/src/scripts/kconfig/menu.c: warning: 'jump' may be used uninitialized in this function [-Wmaybe-uninitialized]: => 587:18
+ /scratch/kisskb/src/sound/core/oss/pcm_oss.c: warning: 'err' may be used uninitialized in this function [-Wuninitialized]: => 2161:6
+ /scratch/kisskb/src/sound/drivers/mpu401/mpu401_uart.c: warning: 'err' may be used uninitialized in this function [-Wuninitialized]: => 312:6, 290:6
+ /scratch/kisskb/src/sound/isa/cmi8330.c: warning: 'card' may be used uninitialized in this function [-Wuninitialized]: => 642:2
+ /scratch/kisskb/src/sound/isa/opti9xx/opti92x-ad1848.c: warning: 'card' may be used uninitialized in this function [-Wuninitialized]: => 1026:17
+ /scratch/kisskb/src/sound/isa/sb/sb16.c: warning: 'card' may be used uninitialized in this function [-Wuninitialized]: => 515:17
+ /scratch/kisskb/src/sound/soc/codecs/arizona.c: warning: 'bclk' may be used uninitialized in this function [-Wuninitialized]: => 1206:15
+ /scratch/kisskb/src/sound/soc/codecs/arizona.c: warning: 'ratio' may be used uninitialized in this function [-Wuninitialized]: => 1455:34
+ /scratch/kisskb/src/sound/soc/codecs/da7210.c: warning: 'pll_div1' may be used uninitialized in this function [-Wuninitialized]: => 1023:15
+ /scratch/kisskb/src/sound/soc/codecs/da7210.c: warning: 'pll_div2' may be used uninitialized in this function [-Wuninitialized]: => 1024:15
+ /scratch/kisskb/src/sound/soc/codecs/da7210.c: warning: 'pll_div3' may be used uninitialized in this function [-Wuninitialized]: => 1025:21
+ /scratch/kisskb/src/sound/soc/codecs/da9055.c: warning: 'pll_frac_bot' may be used uninitialized in this function [-Wuninitialized]: => 1317:15
+ /scratch/kisskb/src/sound/soc/codecs/da9055.c: warning: 'pll_frac_top' may be used uninitialized in this function [-Wuninitialized]: => 1316:15
+ /scratch/kisskb/src/sound/soc/codecs/da9055.c: warning: 'pll_integer' may be used uninitialized in this function [-Wuninitialized]: => 1318:15
+ /scratch/kisskb/src/sound/soc/codecs/wm2200.c: warning: 'fratio' may be used uninitialized in this function [-Wuninitialized]: => 1925:32
+ /scratch/kisskb/src/sound/soc/codecs/wm5100.c: warning: 'fratio' may be used uninitialized in this function [-Wuninitialized]: => 1735:32
+ /scratch/kisskb/src/sound/soc/codecs/wm8350.c: warning: 'fll_div.div' may be used uninitialized in this function [-Wuninitialized]: => 1060:28
+ /scratch/kisskb/src/sound/soc/codecs/wm8350.c: warning: 'fll_div.k' may be used uninitialized in this function [-Wuninitialized]: => 1064:15
+ /scratch/kisskb/src/sound/soc/codecs/wm8350.c: warning: 'fll_div.n' may be used uninitialized in this function [-Wuninitialized]: => 1063:13
+ /scratch/kisskb/src/sound/soc/codecs/wm8350.c: warning: 'fll_div.ratio' may be used uninitialized in this function [-Wuninitialized]: => 1069:54
+ /scratch/kisskb/src/sound/soc/codecs/wm8962.c: warning: 'fratio' may be used uninitialized in this function [-Wuninitialized]: => 2794:32
+ /scratch/kisskb/src/sound/soc/codecs/wm8996.c: warning: 'fratio' may be used uninitialized in this function [-Wuninitialized]: => 1986:32
+ /scratch/kisskb/src/sound/soc/codecs/wm_hubs.c: warning: 'reg_l' may be used uninitialized in this function [-Wuninitialized]: => 325:11
+ /scratch/kisskb/src/sound/soc/codecs/wm_hubs.c: warning: 'reg_r' may be used uninitialized in this function [-Wuninitialized]: => 324:11
+ /scratch/kisskb/src/sound/soc/fsl/fsl_ssi.c: warning: comparison of distinct pointer types lacks a cast [enabled by default]: => 972:3
+ /scratch/kisskb/src/sound/soc/fsl/fsl_ssi.c: warning: passing argument 1 of '__div64_32' from incompatible pointer type [enabled by default]: => 972:3
+ /scratch/kisskb/src/sound/soc/fsl/fsl_ssi.c: warning: right shift count >= width of type [enabled by default]: => 972:3
+ /scratch/kisskb/src/sound/soc/soc-dapm.c: warning: 'ret' may be used uninitialized in this function [-Wuninitialized]: => 3385:6
+ <stdin>: warning: #warning syscall sched_getattr not implemented [-Wcpp]: => 1229:2
+ <stdin>: warning: #warning syscall sched_getattr not implemented: => 1229:2
+ <stdin>: warning: #warning syscall sched_setattr not implemented [-Wcpp]: => 1226:2
+ <stdin>: warning: #warning syscall sched_setattr not implemented: => 1226:2
+ warning: (REMOTEPROC && RPMSG) selects VIRTUALIZATION which has unmet direct dependencies (!CPU_LITTLE_ENDIAN): => N/A
+ warning: (USB_OTG_FSM && FSL_USB2_OTG && USB_MV_OTG) selects USB_OTG which has unmet direct dependencies (USB_SUPPORT && USB && PM_RUNTIME): => N/A
+ warning: drivers/built-in.o(.data+0x9e8): Section mismatch in reference from the variable lba_driver to the function .init.text:lba_driver_probe(): => N/A
+ warning: drivers/built-in.o(.data+0xa88): Section mismatch in reference from the variable ccio_driver to the function .init.text:ccio_probe(): => N/A
+ warning: drivers/built-in.o(.data+0xbe4): Section mismatch in reference from the variable dino_driver to the function .init.text:dino_probe(): => N/A
+ warning: drivers/built-in.o(.data+0xcec): Section mismatch in reference from the variable lasi_driver to the function .init.text:lasi_init_chip(): => N/A
+ warning: drivers/built-in.o(.data+0xd4c): Section mismatch in reference from the variable asp_driver to the function .init.text:asp_init_chip(): => N/A
+ warning: drivers/built-in.o(.data+0xdac): Section mismatch in reference from the variable wax_driver to the function .init.text:wax_init_chip(): => N/A
+ warning: drivers/built-in.o(.data+0xe0c): Section mismatch in reference from the variable eisa_driver to the function .init.text:eisa_probe(): => N/A
+ warning: drivers/built-in.o(.data+0xfb4): Section mismatch in reference from the variable superio_driver to the function .init.text:superio_probe(): => N/A
+ warning: drivers/built-in.o(.text+0x2b4674): Section mismatch in reference from the function etrax_init_module() to the function .init.text:etrax_ethernet_init(): => N/A
+ warning: drivers/net/built-in.o(.text+0x1d8c): Section mismatch in reference from the function etrax_init_module() to the function .init.text:etrax_ethernet_init(): => N/A
+ warning: drivers/net/built-in.o(.text+0x3f0ac): Section mismatch in reference from the function etrax_init_module() to the function .init.text:etrax_ethernet_init(): => N/A
+ warning: vmlinux.o(.data+0x3ff88): Section mismatch in reference from the variable lba_driver to the function .init.text:lba_driver_probe(): => N/A
+ warning: vmlinux.o(.data+0x40028): Section mismatch in reference from the variable ccio_driver to the function .init.text:ccio_probe(): => N/A
+ warning: vmlinux.o(.data+0x40184): Section mismatch in reference from the variable dino_driver to the function .init.text:dino_probe(): => N/A
+ warning: vmlinux.o(.data+0x4028c): Section mismatch in reference from the variable lasi_driver to the function .init.text:lasi_init_chip(): => N/A
+ warning: vmlinux.o(.data+0x402ec): Section mismatch in reference from the variable asp_driver to the function .init.text:asp_init_chip(): => N/A
+ warning: vmlinux.o(.data+0x4034c): Section mismatch in reference from the variable wax_driver to the function .init.text:wax_init_chip(): => N/A
+ warning: vmlinux.o(.data+0x403ac): Section mismatch in reference from the variable eisa_driver to the function .init.text:eisa_probe(): => N/A
+ warning: vmlinux.o(.data+0x40554): Section mismatch in reference from the variable superio_driver to the function .init.text:superio_probe(): => N/A
74 improvements:
- /scratch/kisskb/src/arch/sh/kernel/cpu/sh4/../sh3/../../entry-common.S: Warning: overflow in branch to __restore_all; converted into longer instruction sequence: 89 =>
- /scratch/kisskb/src/arch/sh/kernel/cpu/sh4/../sh3/../../entry-common.S: Warning: overflow in branch to syscall_call; converted into longer instruction sequence: 208 =>
- /scratch/kisskb/src/arch/sh/kernel/cpu/sh4/../sh3/../../entry-common.S: Warning: overflow in branch to syscall_exit_work; converted into longer instruction sequence: 373, 376 => 376
- /scratch/kisskb/src/arch/sh/kernel/cpu/sh4/../sh3/../../entry-common.S: Warning: overflow in branch to syscall_trace_entry; converted into longer instruction sequence: 349, 347 =>
- /scratch/kisskb/src/arch/x86/kvm/../../../virt/kvm/vfio.c: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]: 104:22 =>
- /scratch/kisskb/src/drivers/gpio/gpio-mcp23s08.c: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]: 643:10 =>
- /scratch/kisskb/src/drivers/gpu/drm/gma500/backlight.c: warning: 'do_gma_backlight_set' defined but not used [-Wunused-function]: 29:13 =>
- /scratch/kisskb/src/drivers/gpu/drm/i915/i915_debugfs.c: warning: 'val' may be used uninitialized in this function [-Wuninitialized]: 2393:2 =>
- /scratch/kisskb/src/drivers/infiniband/hw/cxgb4/mem.c: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]: 79:25 =>
- /scratch/kisskb/src/drivers/leds/leds-mc13783.c: warning: 'mask' may be used uninitialized in this function [-Wuninitialized]: 121:41, 197:47 => 120:41
- /scratch/kisskb/src/drivers/leds/leds-mc13783.c: warning: 'reg' may be used uninitialized in this function [-Wuninitialized]: 121:17, 197:23 => 120:17
- /scratch/kisskb/src/drivers/leds/leds-pwm.c: warning: unused variable 'node' [-Wunused-variable]: 88:22 =>
- /scratch/kisskb/src/drivers/md/bcache/alloc.c: warning: 'r' may be used uninitialized in this function [-Wuninitialized]: 416:18 =>
- /scratch/kisskb/src/drivers/mfd/max8998.c: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]: 178:10 =>
- /scratch/kisskb/src/drivers/mfd/tps65217.c: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]: 173:13 =>
- /scratch/kisskb/src/drivers/mtd/devices/docg3.c: warning: unused variable 'base' [-Wunused-variable]: 2122:16 =>
- /scratch/kisskb/src/drivers/net/ethernet/hp/hp100.c: warning: 'cleanup_dev' defined but not used [-Wunused-function]: 2825:13 =>
- /scratch/kisskb/src/drivers/net/ethernet/sfc/tx.c: warning: label 'finish_packet' defined but not used [-Wunused-label]: 432:1 =>
- /scratch/kisskb/src/drivers/net/ethernet/smsc/smc91x.c: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]: 1898:7 =>
- /scratch/kisskb/src/drivers/regulator/ab3100.c: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]: 649:4 =>
- /scratch/kisskb/src/drivers/regulator/tps62360-regulator.c: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]: 363:13 =>
- /scratch/kisskb/src/drivers/rtc/rtc-s5m.c: warning: 's5m_rtc_resume' defined but not used [-Wunused-function]: 642:12 =>
- /scratch/kisskb/src/drivers/rtc/rtc-s5m.c: warning: 's5m_rtc_suspend' defined but not used [-Wunused-function]: 653:12 =>
- /scratch/kisskb/src/drivers/scsi/aic7xxx_old.c: warning: 'aic7xxx_chip_reset' defined but not used [-Wunused-function]: 8341:1 =>
- /scratch/kisskb/src/drivers/scsi/aic7xxx_old.c: warning: 'aic7xxx_configure_bugs' defined but not used [-Wunused-function]: 8895:1 =>
- /scratch/kisskb/src/drivers/scsi/aic7xxx_old.c: warning: 'aic7xxx_load_seeprom' defined but not used [-Wunused-function]: 8496:1 =>
- /scratch/kisskb/src/drivers/scsi/aic7xxx_old.c: warning: case value '257' not in enumerated type 'ahc_chip' [-Wswitch]: 8517:5, 7901:5 =>
- /scratch/kisskb/src/drivers/scsi/aic7xxx_old.c: warning: case value '513' not in enumerated type 'ahc_chip' [-Wswitch]: 7898:5, 8510:5 =>
- /scratch/kisskb/src/drivers/scsi/aic7xxx_old.c: warning: unused variable 'sxfrctl1' [-Wunused-variable]: 8983:17 =>
- /scratch/kisskb/src/drivers/scsi/scsi_pm.c: warning: 'scsi_dev_type_resume' defined but not used [-Wunused-function]: 35:12 =>
- /scratch/kisskb/src/drivers/scsi/scsi_pm.c: warning: 'scsi_dev_type_suspend' defined but not used [-Wunused-function]: 19:12 =>
- /scratch/kisskb/src/drivers/staging/dgap/dgap_driver.c: warning: the frame size of 1036 bytes is larger than 1024 bytes [-Wframe-larger-than=]: 975:1 =>
- /scratch/kisskb/src/drivers/staging/dgap/dgap_fep5.c: warning: the frame size of 1044 bytes is larger than 1024 bytes [-Wframe-larger-than=]: 112:1 =>
- /scratch/kisskb/src/drivers/staging/dgap/dgap_tty.c: warning: the frame size of 1056 bytes is larger than 1024 bytes [-Wframe-larger-than=]: 658:1 =>
- /scratch/kisskb/src/drivers/staging/dgap/dgap_tty.c: warning: the frame size of 1068 bytes is larger than 1024 bytes [-Wframe-larger-than=]: 658:1 =>
- /scratch/kisskb/src/drivers/staging/ft1000/ft1000-usb/ft1000_debug.c: warning: 'msgsz' may be used uninitialized in this function [-Wuninitialized]: 632:54 =>
- /scratch/kisskb/src/drivers/staging/lustre/lustre/llite/file.c: warning: 'count' may be used uninitialized in this function [-Wuninitialized]: 987:9, 1046:9 =>
- /scratch/kisskb/src/drivers/staging/rtl8192e/rtllib_softmac.c: warning: value computed is not used [-Wunused-value]: 924:3 =>
- /scratch/kisskb/src/drivers/tty/ehv_bytechan.c: warning: 'ehv_bc_console_init' defined but not used [-Wunused-function]: 335:19 =>
- /scratch/kisskb/src/drivers/tty/ehv_bytechan.c: warning: data definition has no type or storage class [enabled by default]: 363:1 =>
- /scratch/kisskb/src/drivers/tty/ehv_bytechan.c: warning: parameter names (without types) in function declaration [enabled by default]: 363:1 =>
- /scratch/kisskb/src/drivers/tty/serial/8250/8250_dw.c: warning: 'dw8250_resume' defined but not used [-Wunused-function]: 413:12 =>
- /scratch/kisskb/src/drivers/tty/serial/8250/8250_dw.c: warning: 'dw8250_suspend' defined but not used [-Wunused-function]: 404:12 =>
- /scratch/kisskb/src/drivers/tty/serial/sh-sci.c: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]: 560:23 =>
- /scratch/kisskb/src/fs/sysfs/file.c: warning: unused variable 'kobj' [-Wunused-variable]: 610:18 =>
- /scratch/kisskb/src/include/linux/posix_acl.h: warning: control reaches end of non-void function [-Wreturn-type]: 108:1 =>
- /scratch/kisskb/src/kernel/trace/ring_buffer.c: warning: unused variable 'local_clock_stable' [-Wunused-variable]: 2559:8 =>
- /scratch/kisskb/src/lib/vsprintf.c: warning: 'resource_string.isra.10' uses dynamic stack allocation [enabled by default]: 744:1 =>
- /scratch/kisskb/src/net/netfilter/ipvs/ip_vs_sync.c: warning: unused variable 'ret' [-Wunused-variable]: 1640:8 =>
- modpost: WARNING: modpost: Found 4 section mismatch(es).: N/A =>
- warning: (PPC_CELL_NATIVE && 440EPX && 440GRX && 440GX && 440SPe && 460EX && 460SX && APM821xx && 405EX) selects IBM_EMAC_EMAC4 which has unmet direct dependencies (NETDEVICES && ETHERNET && NET_VENDOR_IBM): N/A =>
- warning: (PPC_CELL_NATIVE && 440GX && 460EX && 460SX && APM821xx) selects IBM_EMAC_TAH which has unmet direct dependencies (NETDEVICES && ETHERNET && NET_VENDOR_IBM): N/A =>
- warning: (PPC_CELL_NATIVE && BLUESTONE && CANYONLANDS && GLACIER && EIGER && 440EPX && 440GRX && 440GX && 460SX && 405EX) selects IBM_EMAC_RGMII which has unmet direct dependencies (NETDEVICES && ETHERNET && NET_VENDOR_IBM): N/A =>
- warning: (PPC_CELL_NATIVE && CANYONLANDS && GLACIER && 440EP && 440EPX && 440GRX && 440GP && 440GX && 460SX && 405GP) selects IBM_EMAC_ZMII which has unmet direct dependencies (NETDEVICES && ETHERNET && NET_VENDOR_IBM): N/A =>
- warning: (SPU_FS) selects MEMORY_HOTPLUG which has unmet direct dependencies ((SPARSEMEM || X86_64_ACPI_NUMA) && ARCH_ENABLE_MEMORY_HOTPLUG && (IA64 || X86 || PPC_BOOK3S_64 || SUPERH || S390)): N/A =>
- warning: drivers/built-in.o(.data+0xa08): Section mismatch in reference from the variable lba_driver to the function .init.text:lba_driver_probe(): N/A =>
- warning: drivers/built-in.o(.data+0xaa8): Section mismatch in reference from the variable ccio_driver to the function .init.text:ccio_probe(): N/A =>
- warning: drivers/built-in.o(.data+0xc04): Section mismatch in reference from the variable dino_driver to the function .init.text:dino_probe(): N/A =>
- warning: drivers/built-in.o(.data+0xd0c): Section mismatch in reference from the variable lasi_driver to the function .init.text:lasi_init_chip(): N/A =>
- warning: drivers/built-in.o(.data+0xd6c): Section mismatch in reference from the variable asp_driver to the function .init.text:asp_init_chip(): N/A =>
- warning: drivers/built-in.o(.data+0xdcc): Section mismatch in reference from the variable wax_driver to the function .init.text:wax_init_chip(): N/A =>
- warning: drivers/built-in.o(.data+0xe2c): Section mismatch in reference from the variable eisa_driver to the function .init.text:eisa_probe(): N/A =>
- warning: drivers/built-in.o(.data+0xfd4): Section mismatch in reference from the variable superio_driver to the function .init.text:superio_probe(): N/A =>
- warning: drivers/net/built-in.o(.text+0x1d78): Section mismatch in reference from the function etrax_init_module() to the function .init.text:etrax_ethernet_init(): N/A =>
- warning: drivers/net/built-in.o(.text+0x3ebdc): Section mismatch in reference from the function etrax_init_module() to the function .init.text:etrax_ethernet_init(): N/A =>
- warning: drivers/usb/gadget/g_audio.o(.text.unlikely+0x114): Section mismatch in reference from the function audio_bind_config() to the function .init.text:afunc_bind(): N/A =>
- warning: vmlinux.o(.data+0x404c8): Section mismatch in reference from the variable lba_driver to the function .init.text:lba_driver_probe(): N/A =>
- warning: vmlinux.o(.data+0x40568): Section mismatch in reference from the variable ccio_driver to the function .init.text:ccio_probe(): N/A =>
- warning: vmlinux.o(.data+0x406c4): Section mismatch in reference from the variable dino_driver to the function .init.text:dino_probe(): N/A =>
- warning: vmlinux.o(.data+0x407cc): Section mismatch in reference from the variable lasi_driver to the function .init.text:lasi_init_chip(): N/A =>
- warning: vmlinux.o(.data+0x4082c): Section mismatch in reference from the variable asp_driver to the function .init.text:asp_init_chip(): N/A =>
- warning: vmlinux.o(.data+0x4088c): Section mismatch in reference from the variable wax_driver to the function .init.text:wax_init_chip(): N/A =>
- warning: vmlinux.o(.data+0x408ec): Section mismatch in reference from the variable eisa_driver to the function .init.text:eisa_probe(): N/A =>
- warning: vmlinux.o(.data+0x40a94): Section mismatch in reference from the variable superio_driver to the function .init.text:superio_probe(): N/A =>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: Build regressions/improvements in v3.14-rc3
2014-02-18 9:02 Build regressions/improvements in v3.14-rc3 Geert Uytterhoeven
@ 2014-02-18 9:06 ` Geert Uytterhoeven
2014-02-19 9:52 ` [PATCH] target_core_alua: silence GCC warning Paul Bolle
2014-02-20 22:11 ` Paul Bolle
2 siblings, 0 replies; 31+ messages in thread
From: Geert Uytterhoeven @ 2014-02-18 9:06 UTC (permalink / raw)
To: linux-kernel; +Cc: linuxppc-dev, linux-xtensa, the arch/x86 maintainers
On Tue, Feb 18, 2014 at 10:02 AM, Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
> JFYI, when comparing v3.14-rc3[1] to v3.14-rc2[3], the summaries are:
> - build errors: +17/-6
+ /scratch/kisskb/src/arch/powerpc/include/asm/floppy.h: error:
'isa_bridge_pcidev' undeclared (first use in this function): =>
142:20
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: #error
m8xx_pcmcia: Bad configuration!: => 127:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error:
'PCMCIA_IO_WIN_NO' undeclared here (not in a function): => 220:30
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error:
'PCMCIA_MEM_WIN_NO' undeclared here (not in a function): => 219:32
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error:
'PCMCIA_SOCKETS_NO' undeclared here (not in a function): => 225:34
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: 'irq_map'
undeclared (first use in this function): => 1027:10
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: 's'
undeclared (first use in this function): => 982:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: expected
')' before 'PCMCIA_BOARD_MSG': => 1036:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'hardware_disable'
[-Werror=implicit-function-declaration]: => 1137:3
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'hardware_enable'
[-Werror=implicit-function-declaration]: => 1080:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'mk_int_int_mask'
[-Werror=implicit-function-declaration]: => 748:6
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'socket_get'
[-Werror=implicit-function-declaration]: => 657:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'voltage_set'
[-Werror=implicit-function-declaration]: => 686:2
powerpc-randconfig
+ /scratch/kisskb/src/drivers/gpio/gpio-xtensa.c: Error: unknown
opcode or format name 'read_impwire': => 87
+ /scratch/kisskb/src/drivers/gpio/gpio-xtensa.c: Error: unknown
opcode or format name 'rur.expstate': => 110
+ /scratch/kisskb/src/drivers/gpio/gpio-xtensa.c: Error: unknown
opcode or format name 'wrmsk_expstate': => 124
xtensa-allmodconfig
+ error: kvm_main.c: undefined reference to `__stack_chk_guard': =>
.text+0x6895)
i386-randconfig
> [1] http://kisskb.ellerman.id.au/kisskb/head/7190/ (all 119 configs)
> [3] http://kisskb.ellerman.id.au/kisskb/head/7167/ (all 119 configs)
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: Build regressions/improvements in v3.14-rc3
@ 2014-02-18 9:06 ` Geert Uytterhoeven
0 siblings, 0 replies; 31+ messages in thread
From: Geert Uytterhoeven @ 2014-02-18 9:06 UTC (permalink / raw)
To: linux-kernel; +Cc: linux-xtensa, the arch/x86 maintainers, linuxppc-dev
On Tue, Feb 18, 2014 at 10:02 AM, Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
> JFYI, when comparing v3.14-rc3[1] to v3.14-rc2[3], the summaries are:
> - build errors: +17/-6
+ /scratch/kisskb/src/arch/powerpc/include/asm/floppy.h: error:
'isa_bridge_pcidev' undeclared (first use in this function): =>
142:20
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: #error
m8xx_pcmcia: Bad configuration!: => 127:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error:
'PCMCIA_IO_WIN_NO' undeclared here (not in a function): => 220:30
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error:
'PCMCIA_MEM_WIN_NO' undeclared here (not in a function): => 219:32
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error:
'PCMCIA_SOCKETS_NO' undeclared here (not in a function): => 225:34
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: 'irq_map'
undeclared (first use in this function): => 1027:10
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: 's'
undeclared (first use in this function): => 982:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: expected
')' before 'PCMCIA_BOARD_MSG': => 1036:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'hardware_disable'
[-Werror=implicit-function-declaration]: => 1137:3
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'hardware_enable'
[-Werror=implicit-function-declaration]: => 1080:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'mk_int_int_mask'
[-Werror=implicit-function-declaration]: => 748:6
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'socket_get'
[-Werror=implicit-function-declaration]: => 657:2
+ /scratch/kisskb/src/drivers/pcmcia/m8xx_pcmcia.c: error: implicit
declaration of function 'voltage_set'
[-Werror=implicit-function-declaration]: => 686:2
powerpc-randconfig
+ /scratch/kisskb/src/drivers/gpio/gpio-xtensa.c: Error: unknown
opcode or format name 'read_impwire': => 87
+ /scratch/kisskb/src/drivers/gpio/gpio-xtensa.c: Error: unknown
opcode or format name 'rur.expstate': => 110
+ /scratch/kisskb/src/drivers/gpio/gpio-xtensa.c: Error: unknown
opcode or format name 'wrmsk_expstate': => 124
xtensa-allmodconfig
+ error: kvm_main.c: undefined reference to `__stack_chk_guard': =>
.text+0x6895)
i386-randconfig
> [1] http://kisskb.ellerman.id.au/kisskb/head/7190/ (all 119 configs)
> [3] http://kisskb.ellerman.id.au/kisskb/head/7167/ (all 119 configs)
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH] target_core_alua: silence GCC warning
2014-02-18 9:02 Build regressions/improvements in v3.14-rc3 Geert Uytterhoeven
2014-02-18 9:06 ` Geert Uytterhoeven
@ 2014-02-19 9:52 ` Paul Bolle
2014-02-19 9:59 ` Geert Uytterhoeven
2014-02-20 22:11 ` Paul Bolle
2 siblings, 1 reply; 31+ messages in thread
From: Paul Bolle @ 2014-02-19 9:52 UTC (permalink / raw)
To: Nicholas A. Bellinger
Cc: Geert Uytterhoeven, linux-scsi, target-devel, linux-kernel
On Tue, 2014-02-18 at 10:02 +0100, Geert Uytterhoeven wrote:
> *** WARNINGS ***
>
> 188 regressions:
> [...]
> + /scratch/kisskb/src/drivers/target/target_core_alua.c: warning: 'alua_ascq' may be used uninitialized in this function [-Wuninitialized]: => 773:18
This one popped up on my (Fedora 20 based) builds on 32 and 64 bits x86.
Geert seems to have seen this on v3.12 already. I first saw it in
v3.14-rc1. Not clear why that is. Anyhow, this is what I came up with to
make this warning go away. (Compile tested only. I don't use this driver
myself.)
-------->8--------
From: Paul Bolle <pebolle@tiscali.nl>
Building target_core_alua.o triggers a GCC warning:
drivers/target/target_core_alua.c: In function ‘target_alua_state_check’:
drivers/target/target_core_alua.c:773:18: warning: ‘alua_ascq’ may be used uninitialized in this function [-Wmaybe-uninitialized]
cmd->scsi_ascq = alua_ascq;
^
This is a false positive. A little trial and error shows it is
apparently caused by core_alua_state_lba_dependent(). It must be hard
for GCC to track the branches of switch statement, inside a
list_for_each_entry loop, inside a while loop.
But if we add a small (inline) helper function we can reorganize the
code a bit. That also allows to drop alua_ascq which, obviously, gets
rid of this warning.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
---
drivers/target/target_core_alua.c | 94 ++++++++++++++++++---------------------
1 file changed, 44 insertions(+), 50 deletions(-)
diff --git a/drivers/target/target_core_alua.c b/drivers/target/target_core_alua.c
index c3d9df6..4111f43 100644
--- a/drivers/target/target_core_alua.c
+++ b/drivers/target/target_core_alua.c
@@ -455,11 +455,26 @@ out:
return rc;
}
+static inline void set_ascq(struct se_cmd *cmd, u8 alua_ascq)
+{
+ /*
+ * Set SCSI additional sense code (ASC) to 'LUN Not Accessible';
+ * The ALUA additional sense code qualifier (ASCQ) is determined
+ * by the ALUA primary or secondary access state..
+ */
+ pr_debug("[%s]: ALUA TG Port not available, "
+ "SenseKey: NOT_READY, ASC/ASCQ: "
+ "0x04/0x%02x\n",
+ cmd->se_tfo->get_fabric_name(), alua_ascq);
+
+ cmd->scsi_asc = 0x04;
+ cmd->scsi_ascq = alua_ascq;
+}
+
static inline int core_alua_state_nonoptimized(
struct se_cmd *cmd,
unsigned char *cdb,
- int nonop_delay_msecs,
- u8 *alua_ascq)
+ int nonop_delay_msecs)
{
/*
* Set SCF_ALUA_NON_OPTIMIZED here, this value will be checked
@@ -471,10 +486,9 @@ static inline int core_alua_state_nonoptimized(
return 0;
}
-static inline int core_alua_state_lba_dependent(
+static noinline int core_alua_state_lba_dependent(
struct se_cmd *cmd,
- struct t10_alua_tg_pt_gp *tg_pt_gp,
- u8 *alua_ascq)
+ struct t10_alua_tg_pt_gp *tg_pt_gp)
{
struct se_device *dev = cmd->se_dev;
u64 segment_size, segment_mult, sectors, lba;
@@ -520,7 +534,7 @@ static inline int core_alua_state_lba_dependent(
}
if (!cur_map) {
spin_unlock(&dev->t10_alua.lba_map_lock);
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
}
list_for_each_entry(map_mem, &cur_map->lba_map_mem_list,
@@ -531,11 +545,11 @@ static inline int core_alua_state_lba_dependent(
switch(map_mem->lba_map_mem_alua_state) {
case ALUA_ACCESS_STATE_STANDBY:
spin_unlock(&dev->t10_alua.lba_map_lock);
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_STANDBY);
return 1;
case ALUA_ACCESS_STATE_UNAVAILABLE:
spin_unlock(&dev->t10_alua.lba_map_lock);
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
default:
break;
@@ -548,8 +562,7 @@ static inline int core_alua_state_lba_dependent(
static inline int core_alua_state_standby(
struct se_cmd *cmd,
- unsigned char *cdb,
- u8 *alua_ascq)
+ unsigned char *cdb)
{
/*
* Allowed CDBs for ALUA_ACCESS_STATE_STANDBY as defined by
@@ -570,7 +583,7 @@ static inline int core_alua_state_standby(
case MI_REPORT_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_STANDBY);
return 1;
}
case MAINTENANCE_OUT:
@@ -578,7 +591,7 @@ static inline int core_alua_state_standby(
case MO_SET_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_STANDBY);
return 1;
}
case REQUEST_SENSE:
@@ -588,7 +601,7 @@ static inline int core_alua_state_standby(
case WRITE_BUFFER:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_STANDBY);
return 1;
}
@@ -597,8 +610,7 @@ static inline int core_alua_state_standby(
static inline int core_alua_state_unavailable(
struct se_cmd *cmd,
- unsigned char *cdb,
- u8 *alua_ascq)
+ unsigned char *cdb)
{
/*
* Allowed CDBs for ALUA_ACCESS_STATE_UNAVAILABLE as defined by
@@ -613,7 +625,7 @@ static inline int core_alua_state_unavailable(
case MI_REPORT_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
}
case MAINTENANCE_OUT:
@@ -621,7 +633,7 @@ static inline int core_alua_state_unavailable(
case MO_SET_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
}
case REQUEST_SENSE:
@@ -629,7 +641,7 @@ static inline int core_alua_state_unavailable(
case WRITE_BUFFER:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
}
@@ -638,8 +650,7 @@ static inline int core_alua_state_unavailable(
static inline int core_alua_state_transition(
struct se_cmd *cmd,
- unsigned char *cdb,
- u8 *alua_ascq)
+ unsigned char *cdb)
{
/*
* Allowed CDBs for ALUA_ACCESS_STATE_TRANSITION as defined by
@@ -654,7 +665,7 @@ static inline int core_alua_state_transition(
case MI_REPORT_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_STATE_TRANSITION;
+ set_ascq(cmd, ASCQ_04H_ALUA_STATE_TRANSITION);
return 1;
}
case REQUEST_SENSE:
@@ -662,7 +673,7 @@ static inline int core_alua_state_transition(
case WRITE_BUFFER:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_STATE_TRANSITION;
+ set_ascq(cmd, ASCQ_04H_ALUA_STATE_TRANSITION);
return 1;
}
@@ -684,8 +695,6 @@ target_alua_state_check(struct se_cmd *cmd)
struct t10_alua_tg_pt_gp *tg_pt_gp;
struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
int out_alua_state, nonop_delay_msecs;
- u8 alua_ascq;
- int ret;
if (dev->se_hba->hba_flags & HBA_FLAGS_INTERNAL_USE)
return 0;
@@ -701,9 +710,8 @@ target_alua_state_check(struct se_cmd *cmd)
if (atomic_read(&port->sep_tg_pt_secondary_offline)) {
pr_debug("ALUA: Got secondary offline status for local"
" target port\n");
- alua_ascq = ASCQ_04H_ALUA_OFFLINE;
- ret = 1;
- goto out;
+ set_ascq(cmd, ASCQ_04H_ALUA_OFFLINE);
+ return TCM_CHECK_CONDITION_NOT_READY;
}
/*
* Second, obtain the struct t10_alua_tg_pt_gp_member pointer to the
@@ -731,20 +739,23 @@ target_alua_state_check(struct se_cmd *cmd)
switch (out_alua_state) {
case ALUA_ACCESS_STATE_ACTIVE_NON_OPTIMIZED:
- ret = core_alua_state_nonoptimized(cmd, cdb,
- nonop_delay_msecs, &alua_ascq);
+ core_alua_state_nonoptimized(cmd, cdb, nonop_delay_msecs);
break;
case ALUA_ACCESS_STATE_STANDBY:
- ret = core_alua_state_standby(cmd, cdb, &alua_ascq);
+ if (core_alua_state_standby(cmd, cdb))
+ return TCM_CHECK_CONDITION_NOT_READY;
break;
case ALUA_ACCESS_STATE_UNAVAILABLE:
- ret = core_alua_state_unavailable(cmd, cdb, &alua_ascq);
+ if (core_alua_state_unavailable(cmd, cdb))
+ return TCM_CHECK_CONDITION_NOT_READY;
break;
case ALUA_ACCESS_STATE_TRANSITION:
- ret = core_alua_state_transition(cmd, cdb, &alua_ascq);
+ if (core_alua_state_transition(cmd, cdb))
+ return TCM_CHECK_CONDITION_NOT_READY;
break;
case ALUA_ACCESS_STATE_LBA_DEPENDENT:
- ret = core_alua_state_lba_dependent(cmd, tg_pt_gp, &alua_ascq);
+ if (core_alua_state_lba_dependent(cmd, tg_pt_gp))
+ return TCM_CHECK_CONDITION_NOT_READY;
break;
/*
* OFFLINE is a secondary ALUA target port group access state, that is
@@ -757,23 +768,6 @@ target_alua_state_check(struct se_cmd *cmd)
return TCM_INVALID_CDB_FIELD;
}
-out:
- if (ret > 0) {
- /*
- * Set SCSI additional sense code (ASC) to 'LUN Not Accessible';
- * The ALUA additional sense code qualifier (ASCQ) is determined
- * by the ALUA primary or secondary access state..
- */
- pr_debug("[%s]: ALUA TG Port not available, "
- "SenseKey: NOT_READY, ASC/ASCQ: "
- "0x04/0x%02x\n",
- cmd->se_tfo->get_fabric_name(), alua_ascq);
-
- cmd->scsi_asc = 0x04;
- cmd->scsi_ascq = alua_ascq;
- return TCM_CHECK_CONDITION_NOT_READY;
- }
-
return 0;
}
--
1.8.5.3
^ permalink raw reply related [flat|nested] 31+ messages in thread
* Re: [PATCH] target_core_alua: silence GCC warning
2014-02-19 9:52 ` [PATCH] target_core_alua: silence GCC warning Paul Bolle
@ 2014-02-19 9:59 ` Geert Uytterhoeven
2014-02-19 10:05 ` Paul Bolle
0 siblings, 1 reply; 31+ messages in thread
From: Geert Uytterhoeven @ 2014-02-19 9:59 UTC (permalink / raw)
To: Paul Bolle; +Cc: Nicholas A. Bellinger, scsi, target-devel, linux-kernel
On Wed, Feb 19, 2014 at 10:52 AM, Paul Bolle <pebolle@tiscali.nl> wrote:
> - ret = core_alua_state_nonoptimized(cmd, cdb,
> - nonop_delay_msecs, &alua_ascq);
> + core_alua_state_nonoptimized(cmd, cdb, nonop_delay_msecs);
I suggest making core_alua_state_nonoptimized() return void, too.
Currently it always returns zero.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: [PATCH] target_core_alua: silence GCC warning
2014-02-19 9:59 ` Geert Uytterhoeven
@ 2014-02-19 10:05 ` Paul Bolle
2014-02-19 22:59 ` Nicholas A. Bellinger
0 siblings, 1 reply; 31+ messages in thread
From: Paul Bolle @ 2014-02-19 10:05 UTC (permalink / raw)
To: Geert Uytterhoeven
Cc: Nicholas A. Bellinger, scsi, target-devel, linux-kernel
On Wed, 2014-02-19 at 10:59 +0100, Geert Uytterhoeven wrote:
> On Wed, Feb 19, 2014 at 10:52 AM, Paul Bolle <pebolle@tiscali.nl> wrote:
> > - ret = core_alua_state_nonoptimized(cmd, cdb,
> > - nonop_delay_msecs, &alua_ascq);
> > + core_alua_state_nonoptimized(cmd, cdb, nonop_delay_msecs);
>
> I suggest making core_alua_state_nonoptimized() return void, too.
> Currently it always returns zero.
Good catch. Not sure how I missed that, since I deliberately ignore its
return value here.
I'll wait with sending a v2 until after Nicholas or the other target
developers have given their feedback.
Paul Bolle
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: [PATCH] target_core_alua: silence GCC warning
2014-02-19 10:05 ` Paul Bolle
@ 2014-02-19 22:59 ` Nicholas A. Bellinger
2014-02-20 8:07 ` [PATCH v2] " Paul Bolle
0 siblings, 1 reply; 31+ messages in thread
From: Nicholas A. Bellinger @ 2014-02-19 22:59 UTC (permalink / raw)
To: Paul Bolle; +Cc: Geert Uytterhoeven, scsi, target-devel, linux-kernel
On Wed, 2014-02-19 at 11:05 +0100, Paul Bolle wrote:
> On Wed, 2014-02-19 at 10:59 +0100, Geert Uytterhoeven wrote:
> > On Wed, Feb 19, 2014 at 10:52 AM, Paul Bolle <pebolle@tiscali.nl> wrote:
> > > - ret = core_alua_state_nonoptimized(cmd, cdb,
> > > - nonop_delay_msecs, &alua_ascq);
> > > + core_alua_state_nonoptimized(cmd, cdb, nonop_delay_msecs);
> >
> > I suggest making core_alua_state_nonoptimized() return void, too.
> > Currently it always returns zero.
>
> Good catch. Not sure how I missed that, since I deliberately ignore its
> return value here.
>
> I'll wait with sending a v2 until after Nicholas or the other target
> developers have given their feedback.
>
>
Hi Paul,
The patch is fine with me.. Feel free to post a -v2.
Thanks,
--nab
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] target_core_alua: silence GCC warning
2014-02-19 22:59 ` Nicholas A. Bellinger
@ 2014-02-20 8:07 ` Paul Bolle
2014-02-20 18:33 ` Nicholas A. Bellinger
0 siblings, 1 reply; 31+ messages in thread
From: Paul Bolle @ 2014-02-20 8:07 UTC (permalink / raw)
To: Nicholas A. Bellinger
Cc: Geert Uytterhoeven, linux-scsi, target-devel, linux-kernel
Building target_core_alua.o triggers a GCC warning:
drivers/target/target_core_alua.c: In function ‘target_alua_state_check’:
drivers/target/target_core_alua.c:773:18: warning: ‘alua_ascq’ may be used uninitialized in this function [-Wmaybe-uninitialized]
cmd->scsi_ascq = alua_ascq;
^
This is a false positive. A little trial and error shows it is
apparently caused by core_alua_state_lba_dependent(). It must be hard
for GCC to track the branches of a switch statement, inside a
list_for_each_entry loop, inside a while loop.
But if we add a small (inline) helper function we can reorganize the
code a bit. That also allows to drop alua_ascq which, obviously, gets
rid of this warning.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
---
v2: Make core_alua_state_nonoptimized() return void, as Geert
suggested.
Also keep core_alua_state_lba_dependent() inline. Setting that function
noinline was just a leftover from the trial and error fase, and isn't
needed to make the warning go away. Ie, I was sloppy in v1!
Still compile tested only.
drivers/target/target_core_alua.c | 95 ++++++++++++++++++---------------------
1 file changed, 44 insertions(+), 51 deletions(-)
diff --git a/drivers/target/target_core_alua.c b/drivers/target/target_core_alua.c
index c3d9df6..fcbe612 100644
--- a/drivers/target/target_core_alua.c
+++ b/drivers/target/target_core_alua.c
@@ -455,11 +455,26 @@ out:
return rc;
}
-static inline int core_alua_state_nonoptimized(
+static inline void set_ascq(struct se_cmd *cmd, u8 alua_ascq)
+{
+ /*
+ * Set SCSI additional sense code (ASC) to 'LUN Not Accessible';
+ * The ALUA additional sense code qualifier (ASCQ) is determined
+ * by the ALUA primary or secondary access state..
+ */
+ pr_debug("[%s]: ALUA TG Port not available, "
+ "SenseKey: NOT_READY, ASC/ASCQ: "
+ "0x04/0x%02x\n",
+ cmd->se_tfo->get_fabric_name(), alua_ascq);
+
+ cmd->scsi_asc = 0x04;
+ cmd->scsi_ascq = alua_ascq;
+}
+
+static inline void core_alua_state_nonoptimized(
struct se_cmd *cmd,
unsigned char *cdb,
- int nonop_delay_msecs,
- u8 *alua_ascq)
+ int nonop_delay_msecs)
{
/*
* Set SCF_ALUA_NON_OPTIMIZED here, this value will be checked
@@ -468,13 +483,11 @@ static inline int core_alua_state_nonoptimized(
*/
cmd->se_cmd_flags |= SCF_ALUA_NON_OPTIMIZED;
cmd->alua_nonop_delay = nonop_delay_msecs;
- return 0;
}
static inline int core_alua_state_lba_dependent(
struct se_cmd *cmd,
- struct t10_alua_tg_pt_gp *tg_pt_gp,
- u8 *alua_ascq)
+ struct t10_alua_tg_pt_gp *tg_pt_gp)
{
struct se_device *dev = cmd->se_dev;
u64 segment_size, segment_mult, sectors, lba;
@@ -520,7 +533,7 @@ static inline int core_alua_state_lba_dependent(
}
if (!cur_map) {
spin_unlock(&dev->t10_alua.lba_map_lock);
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
}
list_for_each_entry(map_mem, &cur_map->lba_map_mem_list,
@@ -531,11 +544,11 @@ static inline int core_alua_state_lba_dependent(
switch(map_mem->lba_map_mem_alua_state) {
case ALUA_ACCESS_STATE_STANDBY:
spin_unlock(&dev->t10_alua.lba_map_lock);
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_STANDBY);
return 1;
case ALUA_ACCESS_STATE_UNAVAILABLE:
spin_unlock(&dev->t10_alua.lba_map_lock);
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
default:
break;
@@ -548,8 +561,7 @@ static inline int core_alua_state_lba_dependent(
static inline int core_alua_state_standby(
struct se_cmd *cmd,
- unsigned char *cdb,
- u8 *alua_ascq)
+ unsigned char *cdb)
{
/*
* Allowed CDBs for ALUA_ACCESS_STATE_STANDBY as defined by
@@ -570,7 +582,7 @@ static inline int core_alua_state_standby(
case MI_REPORT_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_STANDBY);
return 1;
}
case MAINTENANCE_OUT:
@@ -578,7 +590,7 @@ static inline int core_alua_state_standby(
case MO_SET_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_STANDBY);
return 1;
}
case REQUEST_SENSE:
@@ -588,7 +600,7 @@ static inline int core_alua_state_standby(
case WRITE_BUFFER:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_STANDBY);
return 1;
}
@@ -597,8 +609,7 @@ static inline int core_alua_state_standby(
static inline int core_alua_state_unavailable(
struct se_cmd *cmd,
- unsigned char *cdb,
- u8 *alua_ascq)
+ unsigned char *cdb)
{
/*
* Allowed CDBs for ALUA_ACCESS_STATE_UNAVAILABLE as defined by
@@ -613,7 +624,7 @@ static inline int core_alua_state_unavailable(
case MI_REPORT_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
}
case MAINTENANCE_OUT:
@@ -621,7 +632,7 @@ static inline int core_alua_state_unavailable(
case MO_SET_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
}
case REQUEST_SENSE:
@@ -629,7 +640,7 @@ static inline int core_alua_state_unavailable(
case WRITE_BUFFER:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ set_ascq(cmd, ASCQ_04H_ALUA_TG_PT_UNAVAILABLE);
return 1;
}
@@ -638,8 +649,7 @@ static inline int core_alua_state_unavailable(
static inline int core_alua_state_transition(
struct se_cmd *cmd,
- unsigned char *cdb,
- u8 *alua_ascq)
+ unsigned char *cdb)
{
/*
* Allowed CDBs for ALUA_ACCESS_STATE_TRANSITION as defined by
@@ -654,7 +664,7 @@ static inline int core_alua_state_transition(
case MI_REPORT_TARGET_PGS:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_STATE_TRANSITION;
+ set_ascq(cmd, ASCQ_04H_ALUA_STATE_TRANSITION);
return 1;
}
case REQUEST_SENSE:
@@ -662,7 +672,7 @@ static inline int core_alua_state_transition(
case WRITE_BUFFER:
return 0;
default:
- *alua_ascq = ASCQ_04H_ALUA_STATE_TRANSITION;
+ set_ascq(cmd, ASCQ_04H_ALUA_STATE_TRANSITION);
return 1;
}
@@ -684,8 +694,6 @@ target_alua_state_check(struct se_cmd *cmd)
struct t10_alua_tg_pt_gp *tg_pt_gp;
struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
int out_alua_state, nonop_delay_msecs;
- u8 alua_ascq;
- int ret;
if (dev->se_hba->hba_flags & HBA_FLAGS_INTERNAL_USE)
return 0;
@@ -701,9 +709,8 @@ target_alua_state_check(struct se_cmd *cmd)
if (atomic_read(&port->sep_tg_pt_secondary_offline)) {
pr_debug("ALUA: Got secondary offline status for local"
" target port\n");
- alua_ascq = ASCQ_04H_ALUA_OFFLINE;
- ret = 1;
- goto out;
+ set_ascq(cmd, ASCQ_04H_ALUA_OFFLINE);
+ return TCM_CHECK_CONDITION_NOT_READY;
}
/*
* Second, obtain the struct t10_alua_tg_pt_gp_member pointer to the
@@ -731,20 +738,23 @@ target_alua_state_check(struct se_cmd *cmd)
switch (out_alua_state) {
case ALUA_ACCESS_STATE_ACTIVE_NON_OPTIMIZED:
- ret = core_alua_state_nonoptimized(cmd, cdb,
- nonop_delay_msecs, &alua_ascq);
+ core_alua_state_nonoptimized(cmd, cdb, nonop_delay_msecs);
break;
case ALUA_ACCESS_STATE_STANDBY:
- ret = core_alua_state_standby(cmd, cdb, &alua_ascq);
+ if (core_alua_state_standby(cmd, cdb))
+ return TCM_CHECK_CONDITION_NOT_READY;
break;
case ALUA_ACCESS_STATE_UNAVAILABLE:
- ret = core_alua_state_unavailable(cmd, cdb, &alua_ascq);
+ if (core_alua_state_unavailable(cmd, cdb))
+ return TCM_CHECK_CONDITION_NOT_READY;
break;
case ALUA_ACCESS_STATE_TRANSITION:
- ret = core_alua_state_transition(cmd, cdb, &alua_ascq);
+ if (core_alua_state_transition(cmd, cdb))
+ return TCM_CHECK_CONDITION_NOT_READY;
break;
case ALUA_ACCESS_STATE_LBA_DEPENDENT:
- ret = core_alua_state_lba_dependent(cmd, tg_pt_gp, &alua_ascq);
+ if (core_alua_state_lba_dependent(cmd, tg_pt_gp))
+ return TCM_CHECK_CONDITION_NOT_READY;
break;
/*
* OFFLINE is a secondary ALUA target port group access state, that is
@@ -757,23 +767,6 @@ target_alua_state_check(struct se_cmd *cmd)
return TCM_INVALID_CDB_FIELD;
}
-out:
- if (ret > 0) {
- /*
- * Set SCSI additional sense code (ASC) to 'LUN Not Accessible';
- * The ALUA additional sense code qualifier (ASCQ) is determined
- * by the ALUA primary or secondary access state..
- */
- pr_debug("[%s]: ALUA TG Port not available, "
- "SenseKey: NOT_READY, ASC/ASCQ: "
- "0x04/0x%02x\n",
- cmd->se_tfo->get_fabric_name(), alua_ascq);
-
- cmd->scsi_asc = 0x04;
- cmd->scsi_ascq = alua_ascq;
- return TCM_CHECK_CONDITION_NOT_READY;
- }
-
return 0;
}
--
1.8.5.3
^ permalink raw reply related [flat|nested] 31+ messages in thread
* Re: [PATCH v2] target_core_alua: silence GCC warning
2014-02-20 8:07 ` [PATCH v2] " Paul Bolle
@ 2014-02-20 18:33 ` Nicholas A. Bellinger
0 siblings, 0 replies; 31+ messages in thread
From: Nicholas A. Bellinger @ 2014-02-20 18:33 UTC (permalink / raw)
To: Paul Bolle; +Cc: Geert Uytterhoeven, linux-scsi, target-devel, linux-kernel
On Thu, 2014-02-20 at 09:07 +0100, Paul Bolle wrote:
> Building target_core_alua.o triggers a GCC warning:
> drivers/target/target_core_alua.c: In function ‘target_alua_state_check’:
> drivers/target/target_core_alua.c:773:18: warning: ‘alua_ascq’ may be used uninitialized in this function [-Wmaybe-uninitialized]
> cmd->scsi_ascq = alua_ascq;
> ^
>
> This is a false positive. A little trial and error shows it is
> apparently caused by core_alua_state_lba_dependent(). It must be hard
> for GCC to track the branches of a switch statement, inside a
> list_for_each_entry loop, inside a while loop.
>
> But if we add a small (inline) helper function we can reorganize the
> code a bit. That also allows to drop alua_ascq which, obviously, gets
> rid of this warning.
>
> Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
> ---
> v2: Make core_alua_state_nonoptimized() return void, as Geert
> suggested.
>
> Also keep core_alua_state_lba_dependent() inline. Setting that function
> noinline was just a leftover from the trial and error fase, and isn't
> needed to make the warning go away. Ie, I was sloppy in v1!
>
> Still compile tested only.
>
> drivers/target/target_core_alua.c | 95 ++++++++++++++++++---------------------
> 1 file changed, 44 insertions(+), 51 deletions(-)
>
Applied to target-pending/for-next.
Thanks Paul!
--nab
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH] NVMe: silence GCC warning on 32 bit
2014-02-18 9:02 Build regressions/improvements in v3.14-rc3 Geert Uytterhoeven
@ 2014-02-20 22:11 ` Paul Bolle
2014-02-19 9:52 ` [PATCH] target_core_alua: silence GCC warning Paul Bolle
2014-02-20 22:11 ` Paul Bolle
2 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-02-20 22:11 UTC (permalink / raw)
To: Matthew Wilcox; +Cc: Geert Uytterhoeven, linux-nvme, linux-kernel
On Tue, 2014-02-18 at 10:02 +0100, Geert Uytterhoeven wrote:
> *** WARNINGS ***
>
> 188 regressions:
> [...]
> + /scratch/kisskb/src/drivers/block/nvme-core.c: warning: 'bvprv.bv_len' may be used uninitialized in this function [-Wuninitialized]: => 514:18
> + /scratch/kisskb/src/drivers/block/nvme-core.c: warning: 'bvprv.bv_offset' may be used uninitialized in this function [-Wuninitialized]: => 514:18
> + /scratch/kisskb/src/drivers/block/nvme-core.c: warning: 'bvprv.bv_page' may be used uninitialized in this function [-Wuninitialized]: => 511:17
And these popped up in v3.14-rc1 on 32 bit x86. This patch makes these
warnings go away. Compile tested only (on 32 and 64 bit x86).
Review is appreciated, because the code I'm touching here is far from
obvious to me.
-------->8--------
From: Paul Bolle <pebolle@tiscali.nl>
Building nvme-core.o on 32 bit x86 triggers a rather impressive set of
GCC warnings:
In file included from drivers/block/nvme-core.c:20:0:
drivers/block/nvme-core.c: In function 'nvme_submit_bio_queue':
include/linux/bio.h:154:55: warning: 'bvprv.bv_offset' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_offset' was declared here
struct bio_vec bvec, bvprv;
^
In file included from drivers/block/nvme-core.c:20:0:
include/linux/bio.h:154:55: warning: 'bvprv.bv_len' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_len' was declared here
struct bio_vec bvec, bvprv;
^
In file included from [...]/arch/x86/include/asm/page.h:70:0,
from [...]/arch/x86/include/asm/processor.h:17,
from [...]/arch/x86/include/asm/atomic.h:6,
from include/linux/atomic.h:4,
from include/linux/mutex.h:18,
from include/linux/kernfs.h:13,
from include/linux/sysfs.h:15,
from include/linux/kobject.h:21,
from include/linux/pci.h:28,
from include/linux/nvme.h:23,
from drivers/block/nvme-core.c:19:
include/asm-generic/memory_model.h:31:53: warning: 'bvprv.bv_page' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define __page_to_pfn(page) ((unsigned long)((page) - mem_map) + \
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_page' was declared here
struct bio_vec bvec, bvprv;
^
These are false positives. A bit of staring at the code reveals that
"struct bio_vec bvprv" and "int first" operate in lockstep: if first is
1 bvprv isn't yet initialized and if first is 0 bvprv will be
initialized. But if we convert bvprv to a pointer and initialize it to
NULL we can do away with first. And it turns out the warning is gone if
we do that. So that appears to be enough to help GCC understand the
flow of this code.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
---
drivers/block/nvme-core.c | 10 ++++------
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index 51824d1..f9fb28b 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -495,11 +495,10 @@ static int nvme_split_and_submit(struct bio *bio, struct nvme_queue *nvmeq,
static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
struct bio *bio, enum dma_data_direction dma_dir, int psegs)
{
- struct bio_vec bvec, bvprv;
+ struct bio_vec bvec, *bvprv = NULL;
struct bvec_iter iter;
struct scatterlist *sg = NULL;
int length = 0, nsegs = 0, split_len = bio->bi_iter.bi_size;
- int first = 1;
if (nvmeq->dev->stripe_size)
split_len = nvmeq->dev->stripe_size -
@@ -508,10 +507,10 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
sg_init_table(iod->sg, psegs);
bio_for_each_segment(bvec, bio, iter) {
- if (!first && BIOVEC_PHYS_MERGEABLE(&bvprv, &bvec)) {
+ if (bvprv && BIOVEC_PHYS_MERGEABLE(bvprv, &bvec)) {
sg->length += bvec.bv_len;
} else {
- if (!first && BIOVEC_NOT_VIRT_MERGEABLE(&bvprv, &bvec))
+ if (bvprv && BIOVEC_NOT_VIRT_MERGEABLE(bvprv, &bvec))
return nvme_split_and_submit(bio, nvmeq,
length);
@@ -524,8 +523,7 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
if (split_len - length < bvec.bv_len)
return nvme_split_and_submit(bio, nvmeq, split_len);
length += bvec.bv_len;
- bvprv = bvec;
- first = 0;
+ bvprv = &bvec;
}
iod->nents = nsegs;
sg_mark_end(sg);
--
1.8.5.3
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH] NVMe: silence GCC warning on 32 bit
@ 2014-02-20 22:11 ` Paul Bolle
0 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-02-20 22:11 UTC (permalink / raw)
On Tue, 2014-02-18@10:02 +0100, Geert Uytterhoeven wrote:
> *** WARNINGS ***
>
> 188 regressions:
> [...]
> + /scratch/kisskb/src/drivers/block/nvme-core.c: warning: 'bvprv.bv_len' may be used uninitialized in this function [-Wuninitialized]: => 514:18
> + /scratch/kisskb/src/drivers/block/nvme-core.c: warning: 'bvprv.bv_offset' may be used uninitialized in this function [-Wuninitialized]: => 514:18
> + /scratch/kisskb/src/drivers/block/nvme-core.c: warning: 'bvprv.bv_page' may be used uninitialized in this function [-Wuninitialized]: => 511:17
And these popped up in v3.14-rc1 on 32 bit x86. This patch makes these
warnings go away. Compile tested only (on 32 and 64 bit x86).
Review is appreciated, because the code I'm touching here is far from
obvious to me.
-------->8--------
From: Paul Bolle <pebolle@tiscali.nl>
Building nvme-core.o on 32 bit x86 triggers a rather impressive set of
GCC warnings:
In file included from drivers/block/nvme-core.c:20:0:
drivers/block/nvme-core.c: In function 'nvme_submit_bio_queue':
include/linux/bio.h:154:55: warning: 'bvprv.bv_offset' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_offset' was declared here
struct bio_vec bvec, bvprv;
^
In file included from drivers/block/nvme-core.c:20:0:
include/linux/bio.h:154:55: warning: 'bvprv.bv_len' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_len' was declared here
struct bio_vec bvec, bvprv;
^
In file included from [...]/arch/x86/include/asm/page.h:70:0,
from [...]/arch/x86/include/asm/processor.h:17,
from [...]/arch/x86/include/asm/atomic.h:6,
from include/linux/atomic.h:4,
from include/linux/mutex.h:18,
from include/linux/kernfs.h:13,
from include/linux/sysfs.h:15,
from include/linux/kobject.h:21,
from include/linux/pci.h:28,
from include/linux/nvme.h:23,
from drivers/block/nvme-core.c:19:
include/asm-generic/memory_model.h:31:53: warning: 'bvprv.bv_page' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define __page_to_pfn(page) ((unsigned long)((page) - mem_map) + \
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_page' was declared here
struct bio_vec bvec, bvprv;
^
These are false positives. A bit of staring at the code reveals that
"struct bio_vec bvprv" and "int first" operate in lockstep: if first is
1 bvprv isn't yet initialized and if first is 0 bvprv will be
initialized. But if we convert bvprv to a pointer and initialize it to
NULL we can do away with first. And it turns out the warning is gone if
we do that. So that appears to be enough to help GCC understand the
flow of this code.
Signed-off-by: Paul Bolle <pebolle at tiscali.nl>
---
drivers/block/nvme-core.c | 10 ++++------
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index 51824d1..f9fb28b 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -495,11 +495,10 @@ static int nvme_split_and_submit(struct bio *bio, struct nvme_queue *nvmeq,
static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
struct bio *bio, enum dma_data_direction dma_dir, int psegs)
{
- struct bio_vec bvec, bvprv;
+ struct bio_vec bvec, *bvprv = NULL;
struct bvec_iter iter;
struct scatterlist *sg = NULL;
int length = 0, nsegs = 0, split_len = bio->bi_iter.bi_size;
- int first = 1;
if (nvmeq->dev->stripe_size)
split_len = nvmeq->dev->stripe_size -
@@ -508,10 +507,10 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
sg_init_table(iod->sg, psegs);
bio_for_each_segment(bvec, bio, iter) {
- if (!first && BIOVEC_PHYS_MERGEABLE(&bvprv, &bvec)) {
+ if (bvprv && BIOVEC_PHYS_MERGEABLE(bvprv, &bvec)) {
sg->length += bvec.bv_len;
} else {
- if (!first && BIOVEC_NOT_VIRT_MERGEABLE(&bvprv, &bvec))
+ if (bvprv && BIOVEC_NOT_VIRT_MERGEABLE(bvprv, &bvec))
return nvme_split_and_submit(bio, nvmeq,
length);
@@ -524,8 +523,7 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
if (split_len - length < bvec.bv_len)
return nvme_split_and_submit(bio, nvmeq, split_len);
length += bvec.bv_len;
- bvprv = bvec;
- first = 0;
+ bvprv = &bvec;
}
iod->nents = nsegs;
sg_mark_end(sg);
--
1.8.5.3
^ permalink raw reply related [flat|nested] 31+ messages in thread
* Re: [PATCH] NVMe: silence GCC warning on 32 bit
2014-02-20 22:11 ` Paul Bolle
@ 2014-02-21 16:37 ` Keith Busch
-1 siblings, 0 replies; 31+ messages in thread
From: Keith Busch @ 2014-02-21 16:37 UTC (permalink / raw)
To: Paul Bolle; +Cc: Matthew Wilcox, Geert Uytterhoeven, linux-kernel, linux-nvme
On Thu, 20 Feb 2014, Paul Bolle wrote:
> On Tue, 2014-02-18 at 10:02 +0100, Geert Uytterhoeven wrote:
> And these popped up in v3.14-rc1 on 32 bit x86. This patch makes these
> warnings go away. Compile tested only (on 32 and 64 bit x86).
>
> Review is appreciated, because the code I'm touching here is far from
> obvious to me.
> -------->8--------
> From: Paul Bolle <pebolle@tiscali.nl>
> These are false positives. A bit of staring at the code reveals that
> "struct bio_vec bvprv" and "int first" operate in lockstep: if first is
> 1 bvprv isn't yet initialized and if first is 0 bvprv will be
> initialized. But if we convert bvprv to a pointer and initialize it to
> NULL we can do away with first. And it turns out the warning is gone if
> we do that. So that appears to be enough to help GCC understand the
> flow of this code.
That's pretty much how it was done before the bio_vec iterators were
merged, but I think there's a problem with this approach for this patch
(see below).
>
> Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
> ---
> drivers/block/nvme-core.c | 10 ++++------
> 1 file changed, 4 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
> index 51824d1..f9fb28b 100644
> --- a/drivers/block/nvme-core.c
> +++ b/drivers/block/nvme-core.c
> @@ -495,11 +495,10 @@ static int nvme_split_and_submit(struct bio *bio, struct nvme_queue *nvmeq,
> static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
> struct bio *bio, enum dma_data_direction dma_dir, int psegs)
> {
> - struct bio_vec bvec, bvprv;
> + struct bio_vec bvec, *bvprv = NULL;
> struct bvec_iter iter;
> struct scatterlist *sg = NULL;
> int length = 0, nsegs = 0, split_len = bio->bi_iter.bi_size;
> - int first = 1;
>
> if (nvmeq->dev->stripe_size)
> split_len = nvmeq->dev->stripe_size -
> @@ -508,10 +507,10 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
>
> sg_init_table(iod->sg, psegs);
> bio_for_each_segment(bvec, bio, iter) {
> - if (!first && BIOVEC_PHYS_MERGEABLE(&bvprv, &bvec)) {
> + if (bvprv && BIOVEC_PHYS_MERGEABLE(bvprv, &bvec)) {
> sg->length += bvec.bv_len;
> } else {
> - if (!first && BIOVEC_NOT_VIRT_MERGEABLE(&bvprv, &bvec))
> + if (bvprv && BIOVEC_NOT_VIRT_MERGEABLE(bvprv, &bvec))
> return nvme_split_and_submit(bio, nvmeq,
> length);
>
> @@ -524,8 +523,7 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
> if (split_len - length < bvec.bv_len)
> return nvme_split_and_submit(bio, nvmeq, split_len);
> length += bvec.bv_len;
> - bvprv = bvec;
> - first = 0;
> + bvprv = &bvec;
The address of bvec doesn't change, so bvprv is still going to point
to bvec on the next iteration instead of the previous bio_vec like we
want. When the next iteration gets to this comparison:
> + if (bvprv && BIOVEC_PHYS_MERGEABLE(bvprv, &bvec)) {
both bio_vec's have the same address.
> }
> iod->nents = nsegs;
> sg_mark_end(sg);
> --
> 1.8.5.3
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH] NVMe: silence GCC warning on 32 bit
@ 2014-02-21 16:37 ` Keith Busch
0 siblings, 0 replies; 31+ messages in thread
From: Keith Busch @ 2014-02-21 16:37 UTC (permalink / raw)
On Thu, 20 Feb 2014, Paul Bolle wrote:
> On Tue, 2014-02-18@10:02 +0100, Geert Uytterhoeven wrote:
> And these popped up in v3.14-rc1 on 32 bit x86. This patch makes these
> warnings go away. Compile tested only (on 32 and 64 bit x86).
>
> Review is appreciated, because the code I'm touching here is far from
> obvious to me.
> -------->8--------
> From: Paul Bolle <pebolle at tiscali.nl>
> These are false positives. A bit of staring at the code reveals that
> "struct bio_vec bvprv" and "int first" operate in lockstep: if first is
> 1 bvprv isn't yet initialized and if first is 0 bvprv will be
> initialized. But if we convert bvprv to a pointer and initialize it to
> NULL we can do away with first. And it turns out the warning is gone if
> we do that. So that appears to be enough to help GCC understand the
> flow of this code.
That's pretty much how it was done before the bio_vec iterators were
merged, but I think there's a problem with this approach for this patch
(see below).
>
> Signed-off-by: Paul Bolle <pebolle at tiscali.nl>
> ---
> drivers/block/nvme-core.c | 10 ++++------
> 1 file changed, 4 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
> index 51824d1..f9fb28b 100644
> --- a/drivers/block/nvme-core.c
> +++ b/drivers/block/nvme-core.c
> @@ -495,11 +495,10 @@ static int nvme_split_and_submit(struct bio *bio, struct nvme_queue *nvmeq,
> static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
> struct bio *bio, enum dma_data_direction dma_dir, int psegs)
> {
> - struct bio_vec bvec, bvprv;
> + struct bio_vec bvec, *bvprv = NULL;
> struct bvec_iter iter;
> struct scatterlist *sg = NULL;
> int length = 0, nsegs = 0, split_len = bio->bi_iter.bi_size;
> - int first = 1;
>
> if (nvmeq->dev->stripe_size)
> split_len = nvmeq->dev->stripe_size -
> @@ -508,10 +507,10 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
>
> sg_init_table(iod->sg, psegs);
> bio_for_each_segment(bvec, bio, iter) {
> - if (!first && BIOVEC_PHYS_MERGEABLE(&bvprv, &bvec)) {
> + if (bvprv && BIOVEC_PHYS_MERGEABLE(bvprv, &bvec)) {
> sg->length += bvec.bv_len;
> } else {
> - if (!first && BIOVEC_NOT_VIRT_MERGEABLE(&bvprv, &bvec))
> + if (bvprv && BIOVEC_NOT_VIRT_MERGEABLE(bvprv, &bvec))
> return nvme_split_and_submit(bio, nvmeq,
> length);
>
> @@ -524,8 +523,7 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
> if (split_len - length < bvec.bv_len)
> return nvme_split_and_submit(bio, nvmeq, split_len);
> length += bvec.bv_len;
> - bvprv = bvec;
> - first = 0;
> + bvprv = &bvec;
The address of bvec doesn't change, so bvprv is still going to point
to bvec on the next iteration instead of the previous bio_vec like we
want. When the next iteration gets to this comparison:
> + if (bvprv && BIOVEC_PHYS_MERGEABLE(bvprv, &bvec)) {
both bio_vec's have the same address.
> }
> iod->nents = nsegs;
> sg_mark_end(sg);
> --
> 1.8.5.3
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
2014-02-21 16:37 ` Keith Busch
@ 2014-03-04 9:36 ` Paul Bolle
-1 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-03-04 9:36 UTC (permalink / raw)
To: Matthew Wilcox; +Cc: Geert Uytterhoeven, Keith Busch, linux-nvme, linux-kernel
Building nvme-core.o on 32 bit x86 triggers a rather impressive set of
GCC warnings:
In file included from drivers/block/nvme-core.c:20:0:
drivers/block/nvme-core.c: In function 'nvme_submit_bio_queue':
include/linux/bio.h:154:55: warning: 'bvprv.bv_offset' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_offset' was declared here
struct bio_vec bvec, bvprv;
^
In file included from drivers/block/nvme-core.c:20:0:
include/linux/bio.h:154:55: warning: 'bvprv.bv_len' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_len' was declared here
struct bio_vec bvec, bvprv;
^
In file included from [...]/arch/x86/include/asm/page.h:70:0,
from [...]/arch/x86/include/asm/processor.h:17,
from [...]/arch/x86/include/asm/atomic.h:6,
from include/linux/atomic.h:4,
from include/linux/mutex.h:18,
from include/linux/kernfs.h:13,
from include/linux/sysfs.h:15,
from include/linux/kobject.h:21,
from include/linux/pci.h:28,
from include/linux/nvme.h:23,
from drivers/block/nvme-core.c:19:
include/asm-generic/memory_model.h:31:53: warning: 'bvprv.bv_page' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define __page_to_pfn(page) ((unsigned long)((page) - mem_map) + \
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_page' was declared here
struct bio_vec bvec, bvprv;
^
These are false positives. Apparently GCC can't determine that bvprv
will only be used if first is false and has, therefore, been
initialized. It turned out hard to reorganize the code to help GCC
understand the flow of this code. So take the easy way out and
initialize bvprv to { NULL }.
And, since we're touching this code, make first a bool.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
---
v2: redone, as required by Keith's review. Note that initializing bvprv
to { NULL } is already done twice in block/blk-merge.c, which inspired
me to take the easy way out here.
Also note that it's actually not clear to me why these warnings only
trigger on 32 bit. I guess there's some int/long conversion lurking
somewhere. I haven't found it.
bvprv? Would that mean bio_vec private? But it looks like some temporary
variable, so something like tmp may make more sense. Anyhow, still
compile tested only (on 32 and 64 bit x86).
drivers/block/nvme-core.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index 51824d1..60f98be 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -495,11 +495,11 @@ static int nvme_split_and_submit(struct bio *bio, struct nvme_queue *nvmeq,
static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
struct bio *bio, enum dma_data_direction dma_dir, int psegs)
{
- struct bio_vec bvec, bvprv;
+ struct bio_vec bvec, bvprv = { NULL };
struct bvec_iter iter;
struct scatterlist *sg = NULL;
int length = 0, nsegs = 0, split_len = bio->bi_iter.bi_size;
- int first = 1;
+ bool first = true;
if (nvmeq->dev->stripe_size)
split_len = nvmeq->dev->stripe_size -
@@ -525,7 +525,7 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
return nvme_split_and_submit(bio, nvmeq, split_len);
length += bvec.bv_len;
bvprv = bvec;
- first = 0;
+ first = false;
}
iod->nents = nsegs;
sg_mark_end(sg);
--
1.8.5.3
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
@ 2014-03-04 9:36 ` Paul Bolle
0 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-03-04 9:36 UTC (permalink / raw)
Building nvme-core.o on 32 bit x86 triggers a rather impressive set of
GCC warnings:
In file included from drivers/block/nvme-core.c:20:0:
drivers/block/nvme-core.c: In function 'nvme_submit_bio_queue':
include/linux/bio.h:154:55: warning: 'bvprv.bv_offset' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_offset' was declared here
struct bio_vec bvec, bvprv;
^
In file included from drivers/block/nvme-core.c:20:0:
include/linux/bio.h:154:55: warning: 'bvprv.bv_len' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_len' was declared here
struct bio_vec bvec, bvprv;
^
In file included from [...]/arch/x86/include/asm/page.h:70:0,
from [...]/arch/x86/include/asm/processor.h:17,
from [...]/arch/x86/include/asm/atomic.h:6,
from include/linux/atomic.h:4,
from include/linux/mutex.h:18,
from include/linux/kernfs.h:13,
from include/linux/sysfs.h:15,
from include/linux/kobject.h:21,
from include/linux/pci.h:28,
from include/linux/nvme.h:23,
from drivers/block/nvme-core.c:19:
include/asm-generic/memory_model.h:31:53: warning: 'bvprv.bv_page' may be used uninitialized in this function [-Wmaybe-uninitialized]
#define __page_to_pfn(page) ((unsigned long)((page) - mem_map) + \
^
drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_page' was declared here
struct bio_vec bvec, bvprv;
^
These are false positives. Apparently GCC can't determine that bvprv
will only be used if first is false and has, therefore, been
initialized. It turned out hard to reorganize the code to help GCC
understand the flow of this code. So take the easy way out and
initialize bvprv to { NULL }.
And, since we're touching this code, make first a bool.
Signed-off-by: Paul Bolle <pebolle at tiscali.nl>
---
v2: redone, as required by Keith's review. Note that initializing bvprv
to { NULL } is already done twice in block/blk-merge.c, which inspired
me to take the easy way out here.
Also note that it's actually not clear to me why these warnings only
trigger on 32 bit. I guess there's some int/long conversion lurking
somewhere. I haven't found it.
bvprv? Would that mean bio_vec private? But it looks like some temporary
variable, so something like tmp may make more sense. Anyhow, still
compile tested only (on 32 and 64 bit x86).
drivers/block/nvme-core.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index 51824d1..60f98be 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -495,11 +495,11 @@ static int nvme_split_and_submit(struct bio *bio, struct nvme_queue *nvmeq,
static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
struct bio *bio, enum dma_data_direction dma_dir, int psegs)
{
- struct bio_vec bvec, bvprv;
+ struct bio_vec bvec, bvprv = { NULL };
struct bvec_iter iter;
struct scatterlist *sg = NULL;
int length = 0, nsegs = 0, split_len = bio->bi_iter.bi_size;
- int first = 1;
+ bool first = true;
if (nvmeq->dev->stripe_size)
split_len = nvmeq->dev->stripe_size -
@@ -525,7 +525,7 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
return nvme_split_and_submit(bio, nvmeq, split_len);
length += bvec.bv_len;
bvprv = bvec;
- first = 0;
+ first = false;
}
iod->nents = nsegs;
sg_mark_end(sg);
--
1.8.5.3
^ permalink raw reply related [flat|nested] 31+ messages in thread
* Re: [PATCH v2] NVMe: silence GCC warning on 32 bit
2014-03-04 9:36 ` Paul Bolle
@ 2014-03-05 15:09 ` Keith Busch
-1 siblings, 0 replies; 31+ messages in thread
From: Keith Busch @ 2014-03-05 15:09 UTC (permalink / raw)
To: Paul Bolle
Cc: Matthew Wilcox, Geert Uytterhoeven, Keith Busch, linux-nvme,
linux-kernel
Looks good to me. This won't apply in linux-nvme yet and it may be a
little while before it does, so this might be considered to go upstream
through a different tree if you want this in sooner.
On Tue, 4 Mar 2014, Paul Bolle wrote:
> Building nvme-core.o on 32 bit x86 triggers a rather impressive set of
> GCC warnings:
> In file included from drivers/block/nvme-core.c:20:0:
> drivers/block/nvme-core.c: In function 'nvme_submit_bio_queue':
> include/linux/bio.h:154:55: warning: 'bvprv.bv_offset' may be used uninitialized in this function [-Wmaybe-uninitialized]
> #define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
> ^
> drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_offset' was declared here
> struct bio_vec bvec, bvprv;
> ^
> In file included from drivers/block/nvme-core.c:20:0:
> include/linux/bio.h:154:55: warning: 'bvprv.bv_len' may be used uninitialized in this function [-Wmaybe-uninitialized]
> #define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
> ^
> drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_len' was declared here
> struct bio_vec bvec, bvprv;
> ^
> In file included from [...]/arch/x86/include/asm/page.h:70:0,
> from [...]/arch/x86/include/asm/processor.h:17,
> from [...]/arch/x86/include/asm/atomic.h:6,
> from include/linux/atomic.h:4,
> from include/linux/mutex.h:18,
> from include/linux/kernfs.h:13,
> from include/linux/sysfs.h:15,
> from include/linux/kobject.h:21,
> from include/linux/pci.h:28,
> from include/linux/nvme.h:23,
> from drivers/block/nvme-core.c:19:
> include/asm-generic/memory_model.h:31:53: warning: 'bvprv.bv_page' may be used uninitialized in this function [-Wmaybe-uninitialized]
> #define __page_to_pfn(page) ((unsigned long)((page) - mem_map) + \
> ^
> drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_page' was declared here
> struct bio_vec bvec, bvprv;
> ^
>
> These are false positives. Apparently GCC can't determine that bvprv
> will only be used if first is false and has, therefore, been
> initialized. It turned out hard to reorganize the code to help GCC
> understand the flow of this code. So take the easy way out and
> initialize bvprv to { NULL }.
>
> And, since we're touching this code, make first a bool.
>
> Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
> ---
> v2: redone, as required by Keith's review. Note that initializing bvprv
> to { NULL } is already done twice in block/blk-merge.c, which inspired
> me to take the easy way out here.
>
> Also note that it's actually not clear to me why these warnings only
> trigger on 32 bit. I guess there's some int/long conversion lurking
> somewhere. I haven't found it.
>
> bvprv? Would that mean bio_vec private? But it looks like some temporary
> variable, so something like tmp may make more sense. Anyhow, still
> compile tested only (on 32 and 64 bit x86).
bvprv == bio_vec previous
> drivers/block/nvme-core.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
> index 51824d1..60f98be 100644
> --- a/drivers/block/nvme-core.c
> +++ b/drivers/block/nvme-core.c
> @@ -495,11 +495,11 @@ static int nvme_split_and_submit(struct bio *bio, struct nvme_queue *nvmeq,
> static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
> struct bio *bio, enum dma_data_direction dma_dir, int psegs)
> {
> - struct bio_vec bvec, bvprv;
> + struct bio_vec bvec, bvprv = { NULL };
> struct bvec_iter iter;
> struct scatterlist *sg = NULL;
> int length = 0, nsegs = 0, split_len = bio->bi_iter.bi_size;
> - int first = 1;
> + bool first = true;
>
> if (nvmeq->dev->stripe_size)
> split_len = nvmeq->dev->stripe_size -
> @@ -525,7 +525,7 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
> return nvme_split_and_submit(bio, nvmeq, split_len);
> length += bvec.bv_len;
> bvprv = bvec;
> - first = 0;
> + first = false;
> }
> iod->nents = nsegs;
> sg_mark_end(sg);
> --
> 1.8.5.3
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
@ 2014-03-05 15:09 ` Keith Busch
0 siblings, 0 replies; 31+ messages in thread
From: Keith Busch @ 2014-03-05 15:09 UTC (permalink / raw)
Looks good to me. This won't apply in linux-nvme yet and it may be a
little while before it does, so this might be considered to go upstream
through a different tree if you want this in sooner.
On Tue, 4 Mar 2014, Paul Bolle wrote:
> Building nvme-core.o on 32 bit x86 triggers a rather impressive set of
> GCC warnings:
> In file included from drivers/block/nvme-core.c:20:0:
> drivers/block/nvme-core.c: In function 'nvme_submit_bio_queue':
> include/linux/bio.h:154:55: warning: 'bvprv.bv_offset' may be used uninitialized in this function [-Wmaybe-uninitialized]
> #define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
> ^
> drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_offset' was declared here
> struct bio_vec bvec, bvprv;
> ^
> In file included from drivers/block/nvme-core.c:20:0:
> include/linux/bio.h:154:55: warning: 'bvprv.bv_len' may be used uninitialized in this function [-Wmaybe-uninitialized]
> #define bvec_to_phys(bv) (page_to_phys((bv)->bv_page) + (unsigned long) (bv)->bv_offset)
> ^
> drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_len' was declared here
> struct bio_vec bvec, bvprv;
> ^
> In file included from [...]/arch/x86/include/asm/page.h:70:0,
> from [...]/arch/x86/include/asm/processor.h:17,
> from [...]/arch/x86/include/asm/atomic.h:6,
> from include/linux/atomic.h:4,
> from include/linux/mutex.h:18,
> from include/linux/kernfs.h:13,
> from include/linux/sysfs.h:15,
> from include/linux/kobject.h:21,
> from include/linux/pci.h:28,
> from include/linux/nvme.h:23,
> from drivers/block/nvme-core.c:19:
> include/asm-generic/memory_model.h:31:53: warning: 'bvprv.bv_page' may be used uninitialized in this function [-Wmaybe-uninitialized]
> #define __page_to_pfn(page) ((unsigned long)((page) - mem_map) + \
> ^
> drivers/block/nvme-core.c:498:23: note: 'bvprv.bv_page' was declared here
> struct bio_vec bvec, bvprv;
> ^
>
> These are false positives. Apparently GCC can't determine that bvprv
> will only be used if first is false and has, therefore, been
> initialized. It turned out hard to reorganize the code to help GCC
> understand the flow of this code. So take the easy way out and
> initialize bvprv to { NULL }.
>
> And, since we're touching this code, make first a bool.
>
> Signed-off-by: Paul Bolle <pebolle at tiscali.nl>
> ---
> v2: redone, as required by Keith's review. Note that initializing bvprv
> to { NULL } is already done twice in block/blk-merge.c, which inspired
> me to take the easy way out here.
>
> Also note that it's actually not clear to me why these warnings only
> trigger on 32 bit. I guess there's some int/long conversion lurking
> somewhere. I haven't found it.
>
> bvprv? Would that mean bio_vec private? But it looks like some temporary
> variable, so something like tmp may make more sense. Anyhow, still
> compile tested only (on 32 and 64 bit x86).
bvprv == bio_vec previous
> drivers/block/nvme-core.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
> index 51824d1..60f98be 100644
> --- a/drivers/block/nvme-core.c
> +++ b/drivers/block/nvme-core.c
> @@ -495,11 +495,11 @@ static int nvme_split_and_submit(struct bio *bio, struct nvme_queue *nvmeq,
> static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
> struct bio *bio, enum dma_data_direction dma_dir, int psegs)
> {
> - struct bio_vec bvec, bvprv;
> + struct bio_vec bvec, bvprv = { NULL };
> struct bvec_iter iter;
> struct scatterlist *sg = NULL;
> int length = 0, nsegs = 0, split_len = bio->bi_iter.bi_size;
> - int first = 1;
> + bool first = true;
>
> if (nvmeq->dev->stripe_size)
> split_len = nvmeq->dev->stripe_size -
> @@ -525,7 +525,7 @@ static int nvme_map_bio(struct nvme_queue *nvmeq, struct nvme_iod *iod,
> return nvme_split_and_submit(bio, nvmeq, split_len);
> length += bvec.bv_len;
> bvprv = bvec;
> - first = 0;
> + first = false;
> }
> iod->nents = nsegs;
> sg_mark_end(sg);
> --
> 1.8.5.3
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: [PATCH v2] NVMe: silence GCC warning on 32 bit
2014-03-05 15:09 ` Keith Busch
@ 2014-03-06 9:56 ` Paul Bolle
-1 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-03-06 9:56 UTC (permalink / raw)
To: Keith Busch; +Cc: Matthew Wilcox, Geert Uytterhoeven, linux-nvme, linux-kernel
On Wed, 2014-03-05 at 08:09 -0700, Keith Busch wrote:
> Looks good to me.
Thanks.
> This won't apply in linux-nvme yet and it may be a
> little while before it does, so this might be considered to go upstream
> through a different tree if you want this in sooner.
These warnings popped up in v3.14-rc1. (In v3.13 bvprv was a pointer
initialized to NULL.) This patch is not urgent, but it would still be
nice if the warnings were silenced in v3.14, since they are so noisy. Is
there any tree that queues NVMe fixes to be applied during this release
cycle?
> On Tue, 4 Mar 2014, Paul Bolle wrote:
> > bvprv? Would that mean bio_vec private? But it looks like some temporary
> > variable, so something like tmp may make more sense. Anyhow, still
> > compile tested only (on 32 and 64 bit x86).
>
> bvprv == bio_vec previous
Thanks. So perhaps that variable could be renamed to previous (or its
more common alias prev),
Paul Bolle
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
@ 2014-03-06 9:56 ` Paul Bolle
0 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-03-06 9:56 UTC (permalink / raw)
On Wed, 2014-03-05@08:09 -0700, Keith Busch wrote:
> Looks good to me.
Thanks.
> This won't apply in linux-nvme yet and it may be a
> little while before it does, so this might be considered to go upstream
> through a different tree if you want this in sooner.
These warnings popped up in v3.14-rc1. (In v3.13 bvprv was a pointer
initialized to NULL.) This patch is not urgent, but it would still be
nice if the warnings were silenced in v3.14, since they are so noisy. Is
there any tree that queues NVMe fixes to be applied during this release
cycle?
> On Tue, 4 Mar 2014, Paul Bolle wrote:
> > bvprv? Would that mean bio_vec private? But it looks like some temporary
> > variable, so something like tmp may make more sense. Anyhow, still
> > compile tested only (on 32 and 64 bit x86).
>
> bvprv == bio_vec previous
Thanks. So perhaps that variable could be renamed to previous (or its
more common alias prev),
Paul Bolle
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: [PATCH v2] NVMe: silence GCC warning on 32 bit
2014-03-04 9:36 ` Paul Bolle
@ 2014-03-24 13:11 ` Matthew Wilcox
-1 siblings, 0 replies; 31+ messages in thread
From: Matthew Wilcox @ 2014-03-24 13:11 UTC (permalink / raw)
To: Paul Bolle; +Cc: Geert Uytterhoeven, Keith Busch, linux-nvme, linux-kernel
On Tue, Mar 04, 2014 at 10:36:07AM +0100, Paul Bolle wrote:
> Also note that it's actually not clear to me why these warnings only
> trigger on 32 bit. I guess there's some int/long conversion lurking
> somewhere. I haven't found it.
I bet you're using a different revision of the compiler on 32 and 64 bit.
Different gcc revisions have different abilities to track initialisation.
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
@ 2014-03-24 13:11 ` Matthew Wilcox
0 siblings, 0 replies; 31+ messages in thread
From: Matthew Wilcox @ 2014-03-24 13:11 UTC (permalink / raw)
On Tue, Mar 04, 2014@10:36:07AM +0100, Paul Bolle wrote:
> Also note that it's actually not clear to me why these warnings only
> trigger on 32 bit. I guess there's some int/long conversion lurking
> somewhere. I haven't found it.
I bet you're using a different revision of the compiler on 32 and 64 bit.
Different gcc revisions have different abilities to track initialisation.
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: [PATCH v2] NVMe: silence GCC warning on 32 bit
2014-03-24 13:11 ` Matthew Wilcox
@ 2014-03-24 13:31 ` Matthew Wilcox
-1 siblings, 0 replies; 31+ messages in thread
From: Matthew Wilcox @ 2014-03-24 13:31 UTC (permalink / raw)
To: Paul Bolle; +Cc: Keith Busch, Geert Uytterhoeven, linux-kernel, linux-nvme
On Mon, Mar 24, 2014 at 09:11:04AM -0400, Matthew Wilcox wrote:
> On Tue, Mar 04, 2014 at 10:36:07AM +0100, Paul Bolle wrote:
> > Also note that it's actually not clear to me why these warnings only
> > trigger on 32 bit. I guess there's some int/long conversion lurking
> > somewhere. I haven't found it.
>
> I bet you're using a different revision of the compiler on 32 and 64 bit.
> Different gcc revisions have different abilities to track initialisation.
I should try things myself before opening my big mouth. Weird. Using
gcc-4.8, I see the same thing. Guess I should just apply the patch,
though it feels wrong to be initialising an entire struct just to silence
a bogus compiler warning :-(
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
@ 2014-03-24 13:31 ` Matthew Wilcox
0 siblings, 0 replies; 31+ messages in thread
From: Matthew Wilcox @ 2014-03-24 13:31 UTC (permalink / raw)
On Mon, Mar 24, 2014@09:11:04AM -0400, Matthew Wilcox wrote:
> On Tue, Mar 04, 2014@10:36:07AM +0100, Paul Bolle wrote:
> > Also note that it's actually not clear to me why these warnings only
> > trigger on 32 bit. I guess there's some int/long conversion lurking
> > somewhere. I haven't found it.
>
> I bet you're using a different revision of the compiler on 32 and 64 bit.
> Different gcc revisions have different abilities to track initialisation.
I should try things myself before opening my big mouth. Weird. Using
gcc-4.8, I see the same thing. Guess I should just apply the patch,
though it feels wrong to be initialising an entire struct just to silence
a bogus compiler warning :-(
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: [PATCH v2] NVMe: silence GCC warning on 32 bit
2014-03-24 13:31 ` Matthew Wilcox
@ 2014-03-24 15:36 ` Paul Bolle
-1 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-03-24 15:36 UTC (permalink / raw)
To: Matthew Wilcox; +Cc: Keith Busch, Geert Uytterhoeven, linux-kernel, linux-nvme
On Mon, 2014-03-24 at 09:31 -0400, Matthew Wilcox wrote:
> I should try things myself before opening my big mouth. Weird. Using
> gcc-4.8, I see the same thing. Guess I should just apply the patch,
> though it feels wrong to be initialising an entire struct just to silence
> a bogus compiler warning :-(
I noticed this difference on a 32 bit x86 machine and a 64 bit x86
machine that are both running Fedora 20. They both should be at
gcc-4.8.2 for quite some time now (if I grepped the yum log correctly).
Anyhow, the warning on 32 bit is rather noisy, so I wanted it gone. But
my comments should make clear I'm not really happy with this patch.
And as this is now unlikely to be in time for v3.14, we might decide to
dig deeper. It won't be the first time that a rather small change (say,
converting a variable from signed to unsigned) turns out be enough to
make GCC understand the flow of the code.
Thanks,
Paul Bolle
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
@ 2014-03-24 15:36 ` Paul Bolle
0 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-03-24 15:36 UTC (permalink / raw)
On Mon, 2014-03-24@09:31 -0400, Matthew Wilcox wrote:
> I should try things myself before opening my big mouth. Weird. Using
> gcc-4.8, I see the same thing. Guess I should just apply the patch,
> though it feels wrong to be initialising an entire struct just to silence
> a bogus compiler warning :-(
I noticed this difference on a 32 bit x86 machine and a 64 bit x86
machine that are both running Fedora 20. They both should be at
gcc-4.8.2 for quite some time now (if I grepped the yum log correctly).
Anyhow, the warning on 32 bit is rather noisy, so I wanted it gone. But
my comments should make clear I'm not really happy with this patch.
And as this is now unlikely to be in time for v3.14, we might decide to
dig deeper. It won't be the first time that a rather small change (say,
converting a variable from signed to unsigned) turns out be enough to
make GCC understand the flow of the code.
Thanks,
Paul Bolle
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: [PATCH v2] NVMe: silence GCC warning on 32 bit
2014-03-24 15:36 ` Paul Bolle
@ 2014-03-24 15:49 ` Geert Uytterhoeven
-1 siblings, 0 replies; 31+ messages in thread
From: Geert Uytterhoeven @ 2014-03-24 15:49 UTC (permalink / raw)
To: Paul Bolle; +Cc: Matthew Wilcox, Keith Busch, linux-kernel, linux-nvme
Hi Paul,
On Mon, Mar 24, 2014 at 4:36 PM, Paul Bolle <pebolle@tiscali.nl> wrote:
> And as this is now unlikely to be in time for v3.14, we might decide to
> dig deeper. It won't be the first time that a rather small change (say,
> converting a variable from signed to unsigned) turns out be enough to
> make GCC understand the flow of the code.
Anything we can do with factoring out the "if (!first)" in both branches?
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
@ 2014-03-24 15:49 ` Geert Uytterhoeven
0 siblings, 0 replies; 31+ messages in thread
From: Geert Uytterhoeven @ 2014-03-24 15:49 UTC (permalink / raw)
Hi Paul,
On Mon, Mar 24, 2014@4:36 PM, Paul Bolle <pebolle@tiscali.nl> wrote:
> And as this is now unlikely to be in time for v3.14, we might decide to
> dig deeper. It won't be the first time that a rather small change (say,
> converting a variable from signed to unsigned) turns out be enough to
> make GCC understand the flow of the code.
Anything we can do with factoring out the "if (!first)" in both branches?
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert at linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: [PATCH v2] NVMe: silence GCC warning on 32 bit
2014-03-24 15:49 ` Geert Uytterhoeven
@ 2014-03-24 15:57 ` Paul Bolle
-1 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-03-24 15:57 UTC (permalink / raw)
To: Geert Uytterhoeven; +Cc: Matthew Wilcox, Keith Busch, linux-kernel, linux-nvme
On Mon, 2014-03-24 at 16:49 +0100, Geert Uytterhoeven wrote:
> On Mon, Mar 24, 2014 at 4:36 PM, Paul Bolle <pebolle@tiscali.nl> wrote:
> > And as this is now unlikely to be in time for v3.14, we might decide to
> > dig deeper. It won't be the first time that a rather small change (say,
> > converting a variable from signed to unsigned) turns out be enough to
> > make GCC understand the flow of the code.
>
> Anything we can do with factoring out the "if (!first)" in both branches?
That was basically my approach in version one: drop "first" all
together. Worked great: the warning was gone. But Keith noted that I
also completely broke the logic of the code. So I decided, after some
further attempts, to take the easy way out and and initialize "bvprv" to
{ NULL }.
Paul Bolle
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
@ 2014-03-24 15:57 ` Paul Bolle
0 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-03-24 15:57 UTC (permalink / raw)
On Mon, 2014-03-24@16:49 +0100, Geert Uytterhoeven wrote:
> On Mon, Mar 24, 2014@4:36 PM, Paul Bolle <pebolle@tiscali.nl> wrote:
> > And as this is now unlikely to be in time for v3.14, we might decide to
> > dig deeper. It won't be the first time that a rather small change (say,
> > converting a variable from signed to unsigned) turns out be enough to
> > make GCC understand the flow of the code.
>
> Anything we can do with factoring out the "if (!first)" in both branches?
That was basically my approach in version one: drop "first" all
together. Worked great: the warning was gone. But Keith noted that I
also completely broke the logic of the code. So I decided, after some
further attempts, to take the easy way out and and initialize "bvprv" to
{ NULL }.
Paul Bolle
^ permalink raw reply [flat|nested] 31+ messages in thread
* Re: [PATCH v2] NVMe: silence GCC warning on 32 bit
2014-03-24 15:36 ` Paul Bolle
@ 2014-05-08 7:12 ` Paul Bolle
-1 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-05-08 7:12 UTC (permalink / raw)
To: Matthew Wilcox; +Cc: Keith Busch, Geert Uytterhoeven, linux-kernel, linux-nvme
Matthew,
Paul Bolle schreef op ma 24-03-2014 om 16:36 [+0100]:
> On Mon, 2014-03-24 at 09:31 -0400, Matthew Wilcox wrote:
> > I should try things myself before opening my big mouth. Weird. Using
> > gcc-4.8, I see the same thing. Guess I should just apply the patch,
> > though it feels wrong to be initialising an entire struct just to silence
> > a bogus compiler warning :-(
>
> I noticed this difference on a 32 bit x86 machine and a 64 bit x86
> machine that are both running Fedora 20. They both should be at
> gcc-4.8.2 for quite some time now (if I grepped the yum log correctly).
>
> Anyhow, the warning on 32 bit is rather noisy, so I wanted it gone. But
> my comments should make clear I'm not really happy with this patch.
>
> And as this is now unlikely to be in time for v3.14, we might decide to
> dig deeper. It won't be the first time that a rather small change (say,
> converting a variable from signed to unsigned) turns out be enough to
> make GCC understand the flow of the code.
This noisy warning is still seen when compiling v3.15-rc4 for x86 (32
bit, that is). Assuming this patch is not queued anywhere: is the
unsophisticated approach of my v2 acceptable or would you like me to try
and find the cause of this warning?
Paul Bolle
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH v2] NVMe: silence GCC warning on 32 bit
@ 2014-05-08 7:12 ` Paul Bolle
0 siblings, 0 replies; 31+ messages in thread
From: Paul Bolle @ 2014-05-08 7:12 UTC (permalink / raw)
Matthew,
Paul Bolle schreef op ma 24-03-2014 om 16:36 [+0100]:
> On Mon, 2014-03-24@09:31 -0400, Matthew Wilcox wrote:
> > I should try things myself before opening my big mouth. Weird. Using
> > gcc-4.8, I see the same thing. Guess I should just apply the patch,
> > though it feels wrong to be initialising an entire struct just to silence
> > a bogus compiler warning :-(
>
> I noticed this difference on a 32 bit x86 machine and a 64 bit x86
> machine that are both running Fedora 20. They both should be at
> gcc-4.8.2 for quite some time now (if I grepped the yum log correctly).
>
> Anyhow, the warning on 32 bit is rather noisy, so I wanted it gone. But
> my comments should make clear I'm not really happy with this patch.
>
> And as this is now unlikely to be in time for v3.14, we might decide to
> dig deeper. It won't be the first time that a rather small change (say,
> converting a variable from signed to unsigned) turns out be enough to
> make GCC understand the flow of the code.
This noisy warning is still seen when compiling v3.15-rc4 for x86 (32
bit, that is). Assuming this patch is not queued anywhere: is the
unsophisticated approach of my v2 acceptable or would you like me to try
and find the cause of this warning?
Paul Bolle
^ permalink raw reply [flat|nested] 31+ messages in thread
end of thread, other threads:[~2014-05-08 7:12 UTC | newest]
Thread overview: 31+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-02-18 9:02 Build regressions/improvements in v3.14-rc3 Geert Uytterhoeven
2014-02-18 9:06 ` Geert Uytterhoeven
2014-02-18 9:06 ` Geert Uytterhoeven
2014-02-19 9:52 ` [PATCH] target_core_alua: silence GCC warning Paul Bolle
2014-02-19 9:59 ` Geert Uytterhoeven
2014-02-19 10:05 ` Paul Bolle
2014-02-19 22:59 ` Nicholas A. Bellinger
2014-02-20 8:07 ` [PATCH v2] " Paul Bolle
2014-02-20 18:33 ` Nicholas A. Bellinger
2014-02-20 22:11 ` [PATCH] NVMe: silence GCC warning on 32 bit Paul Bolle
2014-02-20 22:11 ` Paul Bolle
2014-02-21 16:37 ` Keith Busch
2014-02-21 16:37 ` Keith Busch
2014-03-04 9:36 ` [PATCH v2] " Paul Bolle
2014-03-04 9:36 ` Paul Bolle
2014-03-05 15:09 ` Keith Busch
2014-03-05 15:09 ` Keith Busch
2014-03-06 9:56 ` Paul Bolle
2014-03-06 9:56 ` Paul Bolle
2014-03-24 13:11 ` Matthew Wilcox
2014-03-24 13:11 ` Matthew Wilcox
2014-03-24 13:31 ` Matthew Wilcox
2014-03-24 13:31 ` Matthew Wilcox
2014-03-24 15:36 ` Paul Bolle
2014-03-24 15:36 ` Paul Bolle
2014-03-24 15:49 ` Geert Uytterhoeven
2014-03-24 15:49 ` Geert Uytterhoeven
2014-03-24 15:57 ` Paul Bolle
2014-03-24 15:57 ` Paul Bolle
2014-05-08 7:12 ` Paul Bolle
2014-05-08 7:12 ` Paul Bolle
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.