| Message ID | 20190830071200.56169-1-chenzhou10@huawei.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | support reserving crashkernel above 4G on arm64 kdump | expand |
Hi all, Friendly ping... On 2019/8/30 15:11, Chen Zhou wrote: > I am busy with other things, so it was a long time before this version was > released. > > This patch series enable reserving crashkernel above 4G in arm64. > > There are following issues in arm64 kdump: > 1. We use crashkernel=X to reserve crashkernel below 4G, which will fail > when there is no enough low memory. > 2. Currently, crashkernel=Y@X can be used to reserve crashkernel above 4G, > in this case, if swiotlb or DMA buffers are requierd, crash dump kernel > will boot failure because there is no low memory available for allocation. > > To solve these issues, introduce crashkernel=X,low to reserve specified > size low memory. > Crashkernel=X tries to reserve memory for the crash dump kernel under > 4G. If crashkernel=Y,low is specified simultaneously, reserve spcified > size low memory for crash kdump kernel devices firstly and then reserve > memory above 4G. > > When crashkernel is reserved above 4G in memory, that is, crashkernel=X,low > is specified simultaneously, kernel should reserve specified size low memory > for crash dump kernel devices. So there may be two crash kernel regions, one > is below 4G, the other is above 4G. > In order to distinct from the high region and make no effect to the use of > kexec-tools, rename the low region as "Crash kernel (low)", and add DT property > "linux,low-memory-range" to crash dump kernel's dtb to pass the low region. > > Besides, we need to modify kexec-tools: > arm64: kdump: add another DT property to crash dump kernel's dtb(see [1]) > > The previous changes and discussions can be retrieved from: > > Changes since [v5] > - Move reserve_crashkernel_low() into kernel/crash_core.c. > - Delete crashkernel=X,high. > - Modify crashkernel=X,low. > If crashkernel=X,low is specified simultaneously, reserve spcified size low > memory for crash kdump kernel devices firstly and then reserve memory above 4G. > In addition, rename crashk_low_res as "Crash kernel (low)" for arm64, and then > pass to crash dump kernel by DT property "linux,low-memory-range". > - Update Documentation/admin-guide/kdump/kdump.rst. > > Changes since [v4] > - Reimplement memblock_cap_memory_ranges for multiple ranges by Mike. > > Changes since [v3] > - Add memblock_cap_memory_ranges back for multiple ranges. > - Fix some compiling warnings. > > Changes since [v2] > - Split patch "arm64: kdump: support reserving crashkernel above 4G" as > two. Put "move reserve_crashkernel_low() into kexec_core.c" in a separate > patch. > > Changes since [v1]: > - Move common reserve_crashkernel_low() code into kernel/kexec_core.c. > - Remove memblock_cap_memory_ranges() i added in v1 and implement that > in fdt_enforce_memory_region(). > There are at most two crash kernel regions, for two crash kernel regions > case, we cap the memory range [min(regs[*].start), max(regs[*].end)] > and then remove the memory range in the middle. > > [1]: http://lists.infradead.org/pipermail/kexec/2019-August/023569.html > [v1]: https://lkml.org/lkml/2019/4/2/1174 > [v2]: https://lkml.org/lkml/2019/4/9/86 > [v3]: https://lkml.org/lkml/2019/4/9/306 > [v4]: https://lkml.org/lkml/2019/4/15/273 > [v5]: https://lkml.org/lkml/2019/5/6/1360 > > Chen Zhou (4): > x86: kdump: move reserve_crashkernel_low() into crash_core.c > arm64: kdump: reserve crashkenel above 4G for crash dump kernel > arm64: kdump: add memory for devices by DT property, low-memory-range > kdump: update Documentation about crashkernel on arm64 > > Documentation/admin-guide/kdump/kdump.rst | 13 ++++- > Documentation/admin-guide/kernel-parameters.txt | 12 ++++- > arch/arm64/include/asm/kexec.h | 3 ++ > arch/arm64/kernel/setup.c | 8 ++- > arch/arm64/mm/init.c | 61 +++++++++++++++++++++-- > arch/x86/include/asm/kexec.h | 3 ++ > arch/x86/kernel/setup.c | 65 +++---------------------- > include/linux/crash_core.h | 4 ++ > include/linux/kexec.h | 1 - > kernel/crash_core.c | 65 +++++++++++++++++++++++++ > 10 files changed, 168 insertions(+), 67 deletions(-) >
On Wed, Dec 18, 2019 at 10:07:59AM +0800, Chen Zhou wrote:
> Friendly ping...
You broke the build:
https://lore.kernel.org/lkml/201909010744.CDe940pv%lkp@intel.com
https://lore.kernel.org/lkml/201909010704.4m9y2sg7%lkp@intel.com
So I doubt anybody will seriously look at this.
Will
HI SEE INLINE ON A QUESTION : > On Dec 17, 2019, at 8:07 PM, Chen Zhou <chenzhou10@huawei.com> wrote: > > Hi all, > > Friendly ping... > > On 2019/8/30 15:11, Chen Zhou wrote: >> I am busy with other things, so it was a long time before this version was >> released. >> >> This patch series enable reserving crashkernel above 4G in arm64. >> >> There are following issues in arm64 kdump: >> 1. We use crashkernel=X to reserve crashkernel below 4G, which will fail >> when there is no enough low memory. >> 2. Currently, crashkernel=Y@X can be used to reserve crashkernel above 4G, >> in this case, if swiotlb or DMA buffers are requierd, crash dump kernel >> will boot failure because there is no low memory available for allocation. Can you elaborate when the boot failures may fail due to lacking swiotlb or DMA buffers ? Are these related to certain adapters or specific platforms ? I have not seen this when using crashkernel=2024M@35GB . >> >> To solve these issues, introduce crashkernel=X,low to reserve specified >> size low memory. >> Crashkernel=X tries to reserve memory for the crash dump kernel under >> 4G. If crashkernel=Y,low is specified simultaneously, reserve spcified >> size low memory for crash kdump kernel devices firstly and then reserve >> memory above 4G. >> >> When crashkernel is reserved above 4G in memory, that is, crashkernel=X,low >> is specified simultaneously, kernel should reserve specified size low memory >> for crash dump kernel devices. So there may be two crash kernel regions, one >> is below 4G, the other is above 4G. >> In order to distinct from the high region and make no effect to the use of >> kexec-tools, rename the low region as "Crash kernel (low)", and add DT property >> "linux,low-memory-range" to crash dump kernel's dtb to pass the low region. >> >> Besides, we need to modify kexec-tools: >> arm64: kdump: add another DT property to crash dump kernel's dtb(see [1]) >> >> The previous changes and discussions can be retrieved from: >> >> Changes since [v5] >> - Move reserve_crashkernel_low() into kernel/crash_core.c. >> - Delete crashkernel=X,high. >> - Modify crashkernel=X,low. >> If crashkernel=X,low is specified simultaneously, reserve spcified size low >> memory for crash kdump kernel devices firstly and then reserve memory above 4G. >> In addition, rename crashk_low_res as "Crash kernel (low)" for arm64, and then >> pass to crash dump kernel by DT property "linux,low-memory-range". >> - Update Documentation/admin-guide/kdump/kdump.rst. >> >> Changes since [v4] >> - Reimplement memblock_cap_memory_ranges for multiple ranges by Mike. >> >> Changes since [v3] >> - Add memblock_cap_memory_ranges back for multiple ranges. >> - Fix some compiling warnings. >> >> Changes since [v2] >> - Split patch "arm64: kdump: support reserving crashkernel above 4G" as >> two. Put "move reserve_crashkernel_low() into kexec_core.c" in a separate >> patch. >> >> Changes since [v1]: >> - Move common reserve_crashkernel_low() code into kernel/kexec_core.c. >> - Remove memblock_cap_memory_ranges() i added in v1 and implement that >> in fdt_enforce_memory_region(). >> There are at most two crash kernel regions, for two crash kernel regions >> case, we cap the memory range [min(regs[*].start), max(regs[*].end)] >> and then remove the memory range in the middle. >> >> [1]: https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.infradead.org_pipermail_kexec_2019-2DAugust_023569.html&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=9tn9kUBabiuYhVtXauANSDGaISnCnHLYcAUQgsPBFxs&e= >> [v1]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_2_1174&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=F-lM7II2cuMF_sK3b6-QhSbWM3X-pI_WZEs0sZitS7A&e= >> [v2]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_9_86&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=5Y-S6sqMTklHkOQsNtjTX3C7pV05BjKLGhJVfMHEvDs&e= >> [v3]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_9_306&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=cWn4zSRQupaZ3jjz4eDvD-pNkoLyL_hsZoRx4yJoD0c&e= >> [v4]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_15_273&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=Nslk4RJKIyIuT0IoQoolXNjupEDXplPhQQwnTSoXNWE&e= >> [v5]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_5_6_1360&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=HJVAM6sCxV2DnNg5d4pw8WPqtkmQnKvztEmkSIgtQ5M&e= >> >> Chen Zhou (4): >> x86: kdump: move reserve_crashkernel_low() into crash_core.c >> arm64: kdump: reserve crashkenel above 4G for crash dump kernel >> arm64: kdump: add memory for devices by DT property, low-memory-range >> kdump: update Documentation about crashkernel on arm64 >> >> Documentation/admin-guide/kdump/kdump.rst | 13 ++++- >> Documentation/admin-guide/kernel-parameters.txt | 12 ++++- >> arch/arm64/include/asm/kexec.h | 3 ++ >> arch/arm64/kernel/setup.c | 8 ++- >> arch/arm64/mm/init.c | 61 +++++++++++++++++++++-- >> arch/x86/include/asm/kexec.h | 3 ++ >> arch/x86/kernel/setup.c | 65 +++---------------------- >> include/linux/crash_core.h | 4 ++ >> include/linux/kexec.h | 1 - >> kernel/crash_core.c | 65 +++++++++++++++++++++++++ >> 10 files changed, 168 insertions(+), 67 deletions(-) >> > > > _______________________________________________ > kexec mailing list > kexec@lists.infradead.org > https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.infradead.org_mailman_listinfo_kexec&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=XMcFx61B_QPg-FUfG_-t88DKCnGm4grqu6zRguiHYrU&e=
Hi Will On 2019/12/18 17:09, Will Deacon wrote: > On Wed, Dec 18, 2019 at 10:07:59AM +0800, Chen Zhou wrote: >> Friendly ping... > > You broke the build: > > https://lore.kernel.org/lkml/201909010744.CDe940pv%lkp@intel.com > https://lore.kernel.org/lkml/201909010704.4m9y2sg7%lkp@intel.com > > So I doubt anybody will seriously look at this. I was also waiting for other suggestions. I will fix this in next version. > > Will > > . Thanks, Chen Zhou
Hi John, On 2019/12/19 1:18, John Donnelly wrote: > HI > > SEE INLINE ON A QUESTION : > >> On Dec 17, 2019, at 8:07 PM, Chen Zhou <chenzhou10@huawei.com> wrote: >> >> Hi all, >> >> Friendly ping... >> >> On 2019/8/30 15:11, Chen Zhou wrote: >>> I am busy with other things, so it was a long time before this version was >>> released. >>> >>> This patch series enable reserving crashkernel above 4G in arm64. >>> >>> There are following issues in arm64 kdump: >>> 1. We use crashkernel=X to reserve crashkernel below 4G, which will fail >>> when there is no enough low memory. >>> 2. Currently, crashkernel=Y@X can be used to reserve crashkernel above 4G, >>> in this case, if swiotlb or DMA buffers are requierd, crash dump kernel >>> will boot failure because there is no low memory available for allocation. > > > Can you elaborate when the boot failures may fail due to lacking swiotlb or DMA buffers ? Are these related to certain adapters or specific platforms ? > > I have not seen this when using crashkernel=2024M@35GB . > For example, in my environment "Huawei TaiShan 2280", we need to use mpt3sas driver in crash dump kernel for dumping vmcore. mpt3sas driver needs to call dma_pool_alloc to allocate some pages, if there is no DMA buffer, page allocation will fail, which leads to crash dump kernel boot failure, like this: [2019/12/19 9:12:41] [ 12.403501] mpt3sas_cm0: diag reset: SUCCESS [2019/12/19 9:12:41] [ 12.456076] mpt3sas_cm0: reply_post_free pool: dma_pool_alloc failed [2019/12/19 9:12:41] [ 12.462515] pci 0004:48:00.0: can't derive routing for PCI INT A [2019/12/19 9:12:41] [ 12.468761] mpt3sas 0004:49:00.0: PCI INT A: no GSI [2019/12/19 9:12:41] [ 12.476348] mpt3sas_cm0: failure at drivers/scsi/mpt3sas/mpt3sas_scsih.c:10626/_scsih_probe()! [2019/12/19 9:14:38] [ TIME ] Timed out waiting for device dev-di…b3\x2d890a\x2d2ead7df26f48.device. [2019/12/19 9:14:38] [DEPEND] Dependency failed for Initrd Root Device. [2019/12/19 9:14:38] [DEPEND] Dependency failed for /sysroot. [2019/12/19 9:14:38] [DEPEND] Dependency failed for Initrd Root File System. [2019/12/19 9:14:38] [DEPEND] Dependency failed for Reload Configuration from the Real Root. [2019/12/19 9:14:38] [DEPEND] Dependency failed for File System C…40bae-9eb8-46b3-890a-2ead7df26f48. Thanks, Chen Zhou > >>> >>> To solve these issues, introduce crashkernel=X,low to reserve specified >>> size low memory. >>> Crashkernel=X tries to reserve memory for the crash dump kernel under >>> 4G. If crashkernel=Y,low is specified simultaneously, reserve spcified >>> size low memory for crash kdump kernel devices firstly and then reserve >>> memory above 4G. >>> >>> When crashkernel is reserved above 4G in memory, that is, crashkernel=X,low >>> is specified simultaneously, kernel should reserve specified size low memory >>> for crash dump kernel devices. So there may be two crash kernel regions, one >>> is below 4G, the other is above 4G. >>> In order to distinct from the high region and make no effect to the use of >>> kexec-tools, rename the low region as "Crash kernel (low)", and add DT property >>> "linux,low-memory-range" to crash dump kernel's dtb to pass the low region. >>> >>> Besides, we need to modify kexec-tools: >>> arm64: kdump: add another DT property to crash dump kernel's dtb(see [1]) >>> >>> The previous changes and discussions can be retrieved from: >>> >>> Changes since [v5] >>> - Move reserve_crashkernel_low() into kernel/crash_core.c. >>> - Delete crashkernel=X,high. >>> - Modify crashkernel=X,low. >>> If crashkernel=X,low is specified simultaneously, reserve spcified size low >>> memory for crash kdump kernel devices firstly and then reserve memory above 4G. >>> In addition, rename crashk_low_res as "Crash kernel (low)" for arm64, and then >>> pass to crash dump kernel by DT property "linux,low-memory-range". >>> - Update Documentation/admin-guide/kdump/kdump.rst. >>> >>> Changes since [v4] >>> - Reimplement memblock_cap_memory_ranges for multiple ranges by Mike. >>> >>> Changes since [v3] >>> - Add memblock_cap_memory_ranges back for multiple ranges. >>> - Fix some compiling warnings. >>> >>> Changes since [v2] >>> - Split patch "arm64: kdump: support reserving crashkernel above 4G" as >>> two. Put "move reserve_crashkernel_low() into kexec_core.c" in a separate >>> patch. >>> >>> Changes since [v1]: >>> - Move common reserve_crashkernel_low() code into kernel/kexec_core.c. >>> - Remove memblock_cap_memory_ranges() i added in v1 and implement that >>> in fdt_enforce_memory_region(). >>> There are at most two crash kernel regions, for two crash kernel regions >>> case, we cap the memory range [min(regs[*].start), max(regs[*].end)] >>> and then remove the memory range in the middle. >>> >>> [1]: https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.infradead.org_pipermail_kexec_2019-2DAugust_023569.html&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=9tn9kUBabiuYhVtXauANSDGaISnCnHLYcAUQgsPBFxs&e= >>> [v1]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_2_1174&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=F-lM7II2cuMF_sK3b6-QhSbWM3X-pI_WZEs0sZitS7A&e= >>> [v2]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_9_86&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=5Y-S6sqMTklHkOQsNtjTX3C7pV05BjKLGhJVfMHEvDs&e= >>> [v3]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_9_306&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=cWn4zSRQupaZ3jjz4eDvD-pNkoLyL_hsZoRx4yJoD0c&e= >>> [v4]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_15_273&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=Nslk4RJKIyIuT0IoQoolXNjupEDXplPhQQwnTSoXNWE&e= >>> [v5]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_5_6_1360&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=HJVAM6sCxV2DnNg5d4pw8WPqtkmQnKvztEmkSIgtQ5M&e= >>> >>> Chen Zhou (4): >>> x86: kdump: move reserve_crashkernel_low() into crash_core.c >>> arm64: kdump: reserve crashkenel above 4G for crash dump kernel >>> arm64: kdump: add memory for devices by DT property, low-memory-range >>> kdump: update Documentation about crashkernel on arm64 >>> >>> Documentation/admin-guide/kdump/kdump.rst | 13 ++++- >>> Documentation/admin-guide/kernel-parameters.txt | 12 ++++- >>> arch/arm64/include/asm/kexec.h | 3 ++ >>> arch/arm64/kernel/setup.c | 8 ++- >>> arch/arm64/mm/init.c | 61 +++++++++++++++++++++-- >>> arch/x86/include/asm/kexec.h | 3 ++ >>> arch/x86/kernel/setup.c | 65 +++---------------------- >>> include/linux/crash_core.h | 4 ++ >>> include/linux/kexec.h | 1 - >>> kernel/crash_core.c | 65 +++++++++++++++++++++++++ >>> 10 files changed, 168 insertions(+), 67 deletions(-) >>> >> >> >> _______________________________________________ >> kexec mailing list >> kexec@lists.infradead.org >> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.infradead.org_mailman_listinfo_kexec&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=XMcFx61B_QPg-FUfG_-t88DKCnGm4grqu6zRguiHYrU&e= > > > . >
Hi John, On 2019/12/20 2:33, John Donnelly wrote: > > >> On Dec 18, 2019, at 8:56 PM, Chen Zhou <chenzhou10@huawei.com> wrote: >> >> Hi John, >> >> On 2019/12/19 1:18, John Donnelly wrote: >>> HI >>> >>> SEE INLINE ON A QUESTION : >>> >>>> On Dec 17, 2019, at 8:07 PM, Chen Zhou <chenzhou10@huawei.com> wrote: >>>> >>>> Hi all, >>>> >>>> Friendly ping... >>>> >>>> On 2019/8/30 15:11, Chen Zhou wrote: >>>>> I am busy with other things, so it was a long time before this version was >>>>> released. >>>>> >>>>> This patch series enable reserving crashkernel above 4G in arm64. >>>>> >>>>> There are following issues in arm64 kdump: >>>>> 1. We use crashkernel=X to reserve crashkernel below 4G, which will fail >>>>> when there is no enough low memory. >>>>> 2. Currently, crashkernel=Y@X can be used to reserve crashkernel above 4G, >>>>> in this case, if swiotlb or DMA buffers are requierd, crash dump kernel >>>>> will boot failure because there is no low memory available for allocation. >>> >>> >>> Can you elaborate when the boot failures may fail due to lacking swiotlb or DMA buffers ? Are these related to certain adapters or specific platforms ? >>> >>> I have not seen this when using crashkernel=2024M@35GB . >>> >> >> For example, in my environment "Huawei TaiShan 2280", >> we need to use mpt3sas driver in crash dump kernel for dumping vmcore. >> >> mpt3sas driver needs to call dma_pool_alloc to allocate some pages, >> if there is no DMA buffer, page allocation will fail, which leads to crash dump kernel boot failure, >> like this: >> >> [2019/12/19 9:12:41] [ 12.403501] mpt3sas_cm0: diag reset: SUCCESS >> [2019/12/19 9:12:41] [ 12.456076] mpt3sas_cm0: reply_post_free pool: dma_pool_alloc failed >> [2019/12/19 9:12:41] [ 12.462515] pci 0004:48:00.0: can't derive routing for PCI INT A >> [2019/12/19 9:12:41] [ 12.468761] mpt3sas 0004:49:00.0: PCI INT A: no GSI >> [2019/12/19 9:12:41] [ 12.476348] mpt3sas_cm0: failure at drivers/scsi/mpt3sas/mpt3sas_scsih.c:10626/_scsih_probe()! >> [2019/12/19 9:14:38] [ TIME ] Timed out waiting for device dev-di…b3\x2d890a\x2d2ead7df26f48.device. >> [2019/12/19 9:14:38] [DEPEND] Dependency failed for Initrd Root Device. >> [2019/12/19 9:14:38] [DEPEND] Dependency failed for /sysroot. >> [2019/12/19 9:14:38] [DEPEND] Dependency failed for Initrd Root File System. >> [2019/12/19 9:14:38] [DEPEND] Dependency failed for Reload Configuration from the Real Root. >> [2019/12/19 9:14:38] [DEPEND] Dependency failed for File System C…40bae-9eb8-46b3-890a-2ead7df26f48. > > > Thank you for sharing . We are not seeing this issue on a 5.4.0.rc8 ; Like I said in a previous email we can take crash dumps using crashkernel=768M for a “ standard “ small VMcore to local storage : > > 0004:01:00.0 RAID bus controller: Broadcom / LSI MegaRAID SAS-3 3316 [Intruder] (rev 01) > Subsystem: Broadcom / LSI MegaRAID SAS 9361-16i > Kernel driver in use: megaraid_sas > > > What version of the kernel are you using ? 5.5.0.rc1 As I said in the patch series cover-letter, this patch series address following cases/isssues. There are following issues in arm64 kdump: 1. We use crashkernel=X to reserve crashkernel below 4G, which will fail when there is no enough low memory. 2. Currently, crashkernel=Y@X can be used to reserve crashkernel above 4G, in this case, if swiotlb or DMA buffers are required, crash dump kernel will boot failure because there is no low memory available for allocation. From your description, you use crashkernel=768M. There are enough crashkernel below 4G, so crashkernel will be reserved successfully and kdump be ok. （This case has no DMA buffers issue.） or you use crashkernel=2024M@35GB, and this case is ok? If your driver doesn't need DMA buffers(I am not sure about it), kdump will also be ok. If i understand your question and explain clearly? Thanks, Chen Zhou > > >> >> Thanks, >> Chen Zhou >> >>> >>>>> >>>>> To solve these issues, introduce crashkernel=X,low to reserve specified >>>>> size low memory. >>>>> Crashkernel=X tries to reserve memory for the crash dump kernel under >>>>> 4G. If crashkernel=Y,low is specified simultaneously, reserve spcified >>>>> size low memory for crash kdump kernel devices firstly and then reserve >>>>> memory above 4G. >>>>> >>>>> When crashkernel is reserved above 4G in memory, that is, crashkernel=X,low >>>>> is specified simultaneously, kernel should reserve specified size low memory >>>>> for crash dump kernel devices. So there may be two crash kernel regions, one >>>>> is below 4G, the other is above 4G. >>>>> In order to distinct from the high region and make no effect to the use of >>>>> kexec-tools, rename the low region as "Crash kernel (low)", and add DT property >>>>> "linux,low-memory-range" to crash dump kernel's dtb to pass the low region. >>>>> >>>>> Besides, we need to modify kexec-tools: >>>>> arm64: kdump: add another DT property to crash dump kernel's dtb(see [1]) >>>>> >>>>> The previous changes and discussions can be retrieved from: >>>>> >>>>> Changes since [v5] >>>>> - Move reserve_crashkernel_low() into kernel/crash_core.c. >>>>> - Delete crashkernel=X,high. >>>>> - Modify crashkernel=X,low. >>>>> If crashkernel=X,low is specified simultaneously, reserve spcified size low >>>>> memory for crash kdump kernel devices firstly and then reserve memory above 4G. >>>>> In addition, rename crashk_low_res as "Crash kernel (low)" for arm64, and then >>>>> pass to crash dump kernel by DT property "linux,low-memory-range". >>>>> - Update Documentation/admin-guide/kdump/kdump.rst. >>>>> >>>>> Changes since [v4] >>>>> - Reimplement memblock_cap_memory_ranges for multiple ranges by Mike. >>>>> >>>>> Changes since [v3] >>>>> - Add memblock_cap_memory_ranges back for multiple ranges. >>>>> - Fix some compiling warnings. >>>>> >>>>> Changes since [v2] >>>>> - Split patch "arm64: kdump: support reserving crashkernel above 4G" as >>>>> two. Put "move reserve_crashkernel_low() into kexec_core.c" in a separate >>>>> patch. >>>>> >>>>> Changes since [v1]: >>>>> - Move common reserve_crashkernel_low() code into kernel/kexec_core.c. >>>>> - Remove memblock_cap_memory_ranges() i added in v1 and implement that >>>>> in fdt_enforce_memory_region(). >>>>> There are at most two crash kernel regions, for two crash kernel regions >>>>> case, we cap the memory range [min(regs[*].start), max(regs[*].end)] >>>>> and then remove the memory range in the middle. >>>>> >>>>> [1]: https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.infradead.org_pipermail_kexec_2019-2DAugust_023569.html&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=9tn9kUBabiuYhVtXauANSDGaISnCnHLYcAUQgsPBFxs&e= >>>>> [v1]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_2_1174&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=F-lM7II2cuMF_sK3b6-QhSbWM3X-pI_WZEs0sZitS7A&e= >>>>> [v2]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_9_86&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=5Y-S6sqMTklHkOQsNtjTX3C7pV05BjKLGhJVfMHEvDs&e= >>>>> [v3]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_9_306&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=cWn4zSRQupaZ3jjz4eDvD-pNkoLyL_hsZoRx4yJoD0c&e= >>>>> [v4]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_4_15_273&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=Nslk4RJKIyIuT0IoQoolXNjupEDXplPhQQwnTSoXNWE&e= >>>>> [v5]: https://urldefense.proofpoint.com/v2/url?u=https-3A__lkml.org_lkml_2019_5_6_1360&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=HJVAM6sCxV2DnNg5d4pw8WPqtkmQnKvztEmkSIgtQ5M&e= >>>>> >>>>> Chen Zhou (4): >>>>> x86: kdump: move reserve_crashkernel_low() into crash_core.c >>>>> arm64: kdump: reserve crashkenel above 4G for crash dump kernel >>>>> arm64: kdump: add memory for devices by DT property, low-memory-range >>>>> kdump: update Documentation about crashkernel on arm64 >>>>> >>>>> Documentation/admin-guide/kdump/kdump.rst | 13 ++++- >>>>> Documentation/admin-guide/kernel-parameters.txt | 12 ++++- >>>>> arch/arm64/include/asm/kexec.h | 3 ++ >>>>> arch/arm64/kernel/setup.c | 8 ++- >>>>> arch/arm64/mm/init.c | 61 +++++++++++++++++++++-- >>>>> arch/x86/include/asm/kexec.h | 3 ++ >>>>> arch/x86/kernel/setup.c | 65 +++---------------------- >>>>> include/linux/crash_core.h | 4 ++ >>>>> include/linux/kexec.h | 1 - >>>>> kernel/crash_core.c | 65 +++++++++++++++++++++++++ >>>>> 10 files changed, 168 insertions(+), 67 deletions(-) >>>>> >>>> >>>> >>>> _______________________________________________ >>>> kexec mailing list >>>> kexec@lists.infradead.org >>>> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.infradead.org_mailman_listinfo_kexec&d=DwICAg&c=RoP1YumCXCgaWHvlZYR8PZh8Bv7qIrMUB65eapI_JnE&r=t2fPg9D87F7D8jm0_3CG9yoiIKdRg4qc_thBw4bzMhc&m=ZAC6UYbT-3qLR3Dvevd09m6neWWzGWSphuvXXlXow68&s=XMcFx61B_QPg-FUfG_-t88DKCnGm4grqu6zRguiHYrU&e= >>> >>> >>> . > > > . >