Re: [PATCH v2 2/2] kexec: fix error propagation in kexec_image_probe_default()
Philipp Rudo <[email protected]>
| Newsgroups | org.infradead.lists.kexec,dev.linux.lists.loongarch,org.infradead.lists.linux-arm-kernel,org.infradead.lists.linux-riscv,org.kernel.vger.linux-kernel |
|---|---|
| Organization | Red Hat inc. |
| Message-ID | <20260821104026.07b7a31d@rotkaeppchen> |
Hi Pratyush, On Thu, 20 Aug 2026 20:11:14 +0200 Pratyush Yadav <[email protected]> wrote: > On Thu, Aug 20 2026, Philipp Rudo wrote: > > > Hi Mukesh, > > Hi Pratyush, > > > > having this patch makes sense. Personally I'd go with having a > > switch-case rather than individual if-blocks. E.g. > > > > switch (ret) { > > case 0: > > image->fops = *fops; > > return 0; > > case -ENOEXEC: > > continue; > > default: > > return ret; > > } > > > > IMHO that is better readable and easier to extend in case other error > > codes need special handling as well. But that is only my personal > > opinion. > > Well, IMO both look roughly the same so I unless Mukesh prefers your > version more, we can keep it as-is. that's why I've marked it as a personal preference. I don't expect Mukesh to switch to it. Still I thought it is worth mentioning just to give an other perspective. > > > > Independent on which implementation you use. > > > > Reviewed-by: Philipp Rudo <[email protected]> > > Thanks for reviewing! > > One tiny suggestion. Can you please reply in-line the next time around? > It is a bit easier reading this response below the code that it refers > to. Agree, especially with larger, more complex patches. In this case however, I thought that the patch is so small that it wouldn't make much of a difference. Anyway, I'll reply in-line next time in such cases as well. Thanks for your review! Philipp > > > > > > On Wed, 19 Aug 2026 23:17:23 +0530 > > Mukesh Pilaniya <[email protected]> wrote: > > > >> kexec_image_probe_default() overwrites ret with each loader's probe > >> return value and returns whatever the last loader returned when no > >> probe matches. The error code reaching userspace depends on whichever > >> loader happens to be last in kexec_file_loaders[]. > >> > >> Only continue to the next loader when a probe returns -ENOEXEC. > >> Propagate any other error such as -ENOMEM immediately. Return -ENOEXEC > >> only when all loaders have been tried and none matched. > >> > >> Suggested-by: Pratyush Yadav <[email protected]> > >> Signed-off-by: Mukesh Pilaniya <[email protected]> > >> --- > >> kernel/kexec_file.c | 12 +++++++----- > >> 1 file changed, 7 insertions(+), 5 deletions(-) > >> > >> diff --git a/kernel/kexec_file.c b/kernel/kexec_file.c > >> index 59fb9d71e9d8..b3060e984ef3 100644 > >> --- a/kernel/kexec_file.c > >> +++ b/kernel/kexec_file.c > >> @@ -68,17 +68,19 @@ int kexec_image_probe_default(struct kimage *image, void *buf, > >> unsigned long buf_len) > >> { > >> const struct kexec_file_ops * const *fops; > >> - int ret = -ENOEXEC; > >> > >> for (fops = &kexec_file_loaders[0]; *fops && (*fops)->probe; ++fops) { > >> - ret = (*fops)->probe(buf, buf_len); > >> - if (!ret) { > >> + int ret = (*fops)->probe(buf, buf_len); > >> + > >> + if (ret == 0) { > >> image->fops = *fops; > >> - return ret; > >> + return 0; > >> } > >> + if (ret != -ENOEXEC) > >> + return ret; > >> } > >> > >> - return ret; > >> + return -ENOEXEC; > >> } > >> > >> static void *kexec_image_load_default(struct kimage *image) > >> > > >